CN114676883A - Power grid operation management method, device and equipment based on big data and storage medium - Google Patents
Power grid operation management method, device and equipment based on big data and storage medium Download PDFInfo
- Publication number
- CN114676883A CN114676883A CN202210196800.XA CN202210196800A CN114676883A CN 114676883 A CN114676883 A CN 114676883A CN 202210196800 A CN202210196800 A CN 202210196800A CN 114676883 A CN114676883 A CN 114676883A
- Authority
- CN
- China
- Prior art keywords
- load
- data
- sequence
- power
- 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.)
- Pending
Links
- 238000007726 management method Methods 0.000 title claims abstract description 43
- 238000003860 storage Methods 0.000 title claims abstract description 14
- 238000001514 detection method Methods 0.000 claims abstract description 62
- 206010000117 Abnormal behaviour Diseases 0.000 claims abstract description 61
- 230000002159 abnormal effect Effects 0.000 claims abstract description 40
- 230000002776 aggregation Effects 0.000 claims abstract description 31
- 238000004220 aggregation Methods 0.000 claims abstract description 31
- 238000007781 pre-processing Methods 0.000 claims abstract description 30
- 238000004140 cleaning Methods 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims description 20
- 230000008859 change Effects 0.000 claims description 11
- 238000007621 cluster analysis Methods 0.000 claims description 6
- 230000010354 integration Effects 0.000 claims description 6
- 238000013499 data model Methods 0.000 claims description 5
- 238000009499 grossing Methods 0.000 claims description 5
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000007906 compression Methods 0.000 claims description 4
- 230000006835 compression Effects 0.000 claims description 4
- 238000002203 pretreatment Methods 0.000 claims description 4
- 230000008439 repair process Effects 0.000 claims description 4
- 238000000034 method Methods 0.000 abstract description 56
- 230000005611 electricity Effects 0.000 description 44
- 230000006399 behavior Effects 0.000 description 34
- 230000008569 process Effects 0.000 description 10
- 230000005856 abnormality Effects 0.000 description 9
- 238000010276 construction Methods 0.000 description 8
- 238000007418 data mining Methods 0.000 description 8
- 238000010248 power generation Methods 0.000 description 8
- 230000002354 daily effect Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000009826 distribution Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000005065 mining Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000003203 everyday effect Effects 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013501 data transformation Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002688 persistence Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 230000005612 types of electricity Effects 0.000 description 1
Images
Classifications
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Evolutionary Computation (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Strategic Management (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a power grid operation management method, a device, equipment and a storage medium based on big data, wherein the method comprises the following steps: acquiring power load data, and cleaning abnormal data in the power load data by performing preprocessing operation on the power load data; performing density clustering of fluctuation intervals on the power load data subjected to the preprocessing operation to obtain an outlier sequence for abnormal detection and a load sequence cluster for prediction, establishing an abnormal behavior detection model according to the outlier sequence, and establishing a load prediction model according to the load sequence cluster; and performing abnormal behavior detection on the outlier sequence by using an abnormal behavior detection model to obtain an abnormal behavior sequence, performing short-term load prediction on the load sequence aggregation cluster by using a load prediction model to obtain a prediction sequence, and managing the operation of the power grid on the basis of the abnormal behavior sequence and the prediction sequence. Therefore, abnormal power utilization behavior detection and short-term prediction of power load are achieved, and safe and reliable operation of a power grid is guaranteed.
Description
Technical Field
The invention relates to the field of power grid operation management, in particular to a power grid operation management method, a power grid operation management device, power grid operation management equipment and a computer readable storage medium based on big data.
Background
In recent years, the construction of smart cities attracts people's extensive attention, and smart power grids play a role as important components of smart cities and play a considerable role in the construction process of smart cities. The construction of smart cities requires a more intelligent power system, and with the gradual maturity of big data technologies, the big data technologies are applied to solve the problems of the power industry, and especially, the research of mining the power data by using the data mining technology becomes one of the hot spots for smart grid construction. Because the growth speed of the power demand side in China is about to exceed the growth of the generated energy, how to realize the safe and stable operation of a power system becomes the key point of power grid construction. In the operation process of the power system, due to the complex and changeable environment and the rapid development of the society, a series of abnormal power utilization behaviors which influence the stable operation of the power system may occur. The abnormalities not only affect the planning construction of the power grid and the scheduling arrangement of the power, but also mislead the formulation of the regional economic plan and hinder the development of the society. The continuous and reliable operation of the power system is not only related to the daily life of thousands of households and the operation and development of enterprises, but also related to the long-term security of the country. Therefore, ensuring safe and reliable operation of the power system is an important requirement for guaranteeing the national civilization.
Disclosure of Invention
The invention mainly aims to provide a power grid operation management method based on big data, and aims to solve the technical problem of how to realize safe and stable operation of a power system in the prior art.
In order to achieve the above object, the present invention provides a power grid operation management method based on big data, which includes:
acquiring power load data, and cleaning abnormal data in the power load data by performing preprocessing operation on the power load data;
performing density clustering of fluctuation intervals on the power load data subjected to preprocessing operation to obtain an outlier sequence for abnormal detection and a load sequence aggregation cluster for prediction, establishing an abnormal behavior detection model according to the outlier sequence, and establishing a load prediction model according to the load sequence aggregation cluster;
and performing abnormal behavior detection on the outlier sequence by using the abnormal behavior detection model to obtain an abnormal behavior sequence, performing short-term load prediction on the load sequence aggregation cluster by using the load prediction model to obtain a prediction sequence, and managing the operation of the power grid based on the abnormal behavior sequence and the prediction sequence.
Optionally, the anomaly data includes noise data and missing value data,
the step of clearing abnormal data in the power load data comprises the following steps:
and cleaning the noise data and the missing value data according to a preset time sequence model.
Optionally, the step of cleaning the noise data and the missing value data according to a preset time series model includes:
if the cleaned data object is the noise data, after the noise data is detected based on the front and back fluctuation relation of the load sequence and the distance between the loads, adopting the change data of the current load relative to the time sequence of the previous preset time period to repair;
if the cleaned data object is the missing value data, acquiring load data of the same time point of adjacent preset time periods before and after the missing value in the load sequence, calculating according to the load data of the same time point to obtain a load average value, obtaining a load variation based on the load variation rate of the next time period relative to the previous time period, and adding the load variation to fill the missing value data on the basis of the load average value.
Optionally, the step of performing abnormal behavior detection on the outlier sequence by using the abnormal behavior detection model to obtain an abnormal behavior sequence includes:
Reading user information to obtain a load sequence of a user in a preset reading period, performing cluster analysis on the load sequence to obtain a similar category group, and judging whether an outlier sequence exists in the load sequence based on the similar category group;
if the outlier sequence exists, calculating the outlier sequence by using a preset historical data model to obtain a historical matching degree, and calculating the outlier sequence by using a preset similar user model to obtain a similar matching degree;
and obtaining a final matching degree based on the historical matching degree and the similarity matching degree, and performing matching processing of the final matching degree on the outlier sequence to obtain the abnormal behavior sequence.
Optionally, the step of performing short-term prediction of load on the load sequence aggregation cluster by using the load prediction model to obtain a prediction sequence includes:
and judging whether the target date to be predicted in the short term is a vacation or not, if so, extracting a fluctuation sequence and a synchronization sequence of the last year two days before the target date to be predicted in the short term, and outputting a prediction sequence.
Optionally, the step of determining whether the target date to be short-term predicted is after the vacation further comprises:
If the target date to be predicted in the short term is not a holiday, selecting a preset number of load sequences adjacent to the target date from the load sequence aggregation cluster, and judging whether the preset number of load sequences and the load sequences to be predicted in the short term are in the same time period or not;
if the preset number of load sequences and the load sequences to be predicted in a short term are in the same time period, intercepting the load sequences which are away from the target date by preset interval time from the selected load sequences to serve as reference sequences, and outputting the predicted sequences based on the reference sequences;
and if the preset number of load sequences and the load sequences to be predicted in the short term are not in the same time period, removing the load sequences to be predicted in the short term from the load sequence cluster.
Optionally, the step of performing a preprocessing operation on the power load data comprises:
one or more of the following pre-treatment operations are employed,
performing data cleaning on the power load data by filling missing values, smoothing noise data and deleting outliers;
or, performing data integration on the power load data through deduplication processing;
Or, carrying out data specification on the power load data through compression;
or, the power load data is subjected to data conversion processing through data discretization and hierarchical processing.
In addition, in order to achieve the above object, the present invention further provides a big data based power grid operation management apparatus, including:
the preprocessing module is used for acquiring power load data, preprocessing the power load data and clearing abnormal data in the power load data;
the density clustering module is used for performing density clustering of fluctuation intervals on the power load data subjected to the preprocessing operation, and establishing an abnormal behavior detection model and a load prediction model;
and the detection prediction module is used for performing abnormal behavior detection on the outlier sequence by using the abnormal behavior detection model to obtain an abnormal behavior sequence, performing short-term prediction on the load of the load sequence aggregation cluster by using the load prediction model to obtain a prediction sequence, and managing the operation of the power grid based on the abnormal behavior sequence and the prediction sequence.
In addition, to achieve the above object, the present invention further provides a big data-based power grid operation management device, including: the system comprises a memory, a processor and a big data-based power grid operation management program stored on the memory and capable of running on the processor, wherein the big data-based power grid operation management program realizes the steps of the big data-based power grid operation management method when being executed by the processor.
In addition, to achieve the above object, the present invention also provides a computer readable storage medium, on which a big data based power grid operation management program is stored, and when the big data based power grid operation management program is executed by a processor, the computer readable storage medium implements the steps of the big data based power grid operation management method as described above.
In the power grid operation management method, the power grid operation management equipment and the computer-readable storage medium based on the big data, which are provided by the embodiment of the invention, in view of various sources of load data, in order to improve the usability of the data, a method for analyzing the use time sequence of the power data is preprocessed; for the detection of abnormal behaviors, density clustering is carried out on fluctuation intervals of the power loads to obtain outliers, then matching processing is carried out by utilizing historical power utilization information of the user and power utilization load characteristic curves of similar users of the user, and abnormal suspicion degree of the outliers is obtained through analysis; because the power dispatching can not be independent of the prediction of the power load, a density clustering prediction method based on the time sequence is provided according to the prediction demand of the power load and the demand. The method comprises the steps of representing a power utilization mode by indexes such as starting time, fluctuation time span, starting power consumption, average growth rate and the like of a power utilization sequence, clustering the power utilization mode to obtain similar load cluster, and performing short-term prediction on user loads by combining a regression method. During anomaly detection, relevance among similar users is rarely considered in the prior art, and when the power utilization behaviors of users of the same similar type all change, if only the power utilization behavior of a single user is considered, misjudgment may be caused, and the anomaly detection effect is influenced. Therefore, a multi-user abnormal electricity utilization behavior detection method based on density clustering is provided. In the aspect of load prediction, in order to improve the prediction time precision, a method of clustering load fluctuation intervals is adopted to perform short-term prediction on the electric load. Therefore, the detection of abnormal electricity utilization behaviors and the short-term prediction of the power load are realized, and the safe and reliable operation of the power grid is guaranteed.
Drawings
Fig. 1 is a schematic structural diagram of an operating device of a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart illustrating an embodiment of a big data-based power grid operation management method according to the present invention;
FIG. 3 is a flow chart of abnormal load behavior detection based on density clustering according to an embodiment of the big data-based power grid operation management method of the present invention;
fig. 4 is a load prediction flowchart of an embodiment of a big data-based power grid operation management method according to the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a terminal structural diagram of a hardware operating environment according to an embodiment of the present invention.
As shown in fig. 1, the operation device may include: a processor 1001, such as a Central Processing Unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a WIreless interface (e.g., a WIreless-FIdelity (WI-FI) interface). The Memory 1005 may be a Random Access Memory (RAM) Memory, or may be a Non-Volatile Memory (NVM), such as a disk Memory. The memory 1005 may alternatively be a storage device separate from the processor 1001.
Those skilled in the art will appreciate that the configuration shown in FIG. 1 does not constitute a limitation of the operating device and may include more or fewer components than shown, or some components may be combined, or a different arrangement of components.
As shown in fig. 1, the memory 1005, which is a storage medium, may include therein an operating system, a data storage module, a network communication module, a user interface module, and a big data-based power grid operation management program.
In the operating device shown in fig. 1, the network interface 1004 is mainly used for data communication with other devices; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the running device of the present invention may be provided in a running device, and the running device calls the grid running management program based on big data stored in the memory 1005 through the processor 1001 and performs the following operations:
acquiring power load data, and cleaning abnormal data in the power load data by performing preprocessing operation on the power load data;
performing density clustering of fluctuation intervals on the power load data subjected to preprocessing operation to obtain an outlier sequence for abnormal detection and a load sequence aggregation cluster for prediction, establishing an abnormal behavior detection model according to the outlier sequence, and establishing a load prediction model according to the load sequence aggregation cluster;
And performing abnormal behavior detection on the outlier sequence by using the abnormal behavior detection model to obtain an abnormal behavior sequence, performing short-term load prediction on the load sequence aggregation cluster by using the load prediction model to obtain a prediction sequence, and managing the operation of the power grid based on the abnormal behavior sequence and the prediction sequence.
Further, the processor 1001 may call the big data based grid operation management program stored in the memory 1005, and further perform the following operations:
the anomaly data includes noise data and missing value data,
the step of clearing abnormal data in the power load data comprises the following steps:
and cleaning the noise data and the missing value data according to a preset time sequence model.
Further, the processor 1001 may call the big data based grid operation management program stored in the memory 1005, and also perform the following operations:
the step of cleaning the noise data and the missing value data according to a preset time series model comprises:
if the cleaned data object is the noise data, after the noise data is detected based on the front and back fluctuation relation of the load sequence and the distance between the loads, adopting the change data of the current load relative to the time sequence of the previous preset time period to repair;
If the cleaned data object is the missing value data, acquiring load data of the same time point of adjacent preset time periods before and after the missing value in the load sequence, calculating according to the load data of the same time point to obtain a load average value, obtaining a load variation based on the load variation rate of the next time period relative to the previous time period, and adding the load variation to fill the missing value data on the basis of the load average value.
Further, the processor 1001 may call the big data based grid operation management program stored in the memory 1005, and further perform the following operations:
reading user information to obtain a load sequence of a user in a preset reading period, performing cluster analysis on the load sequence to obtain a similar category group, and judging whether an outlier sequence exists in the load sequence based on the similar category group;
if the outlier sequence exists, calculating the outlier sequence by using a preset historical data model to obtain a historical matching degree, and calculating the outlier sequence by using a preset similar user model to obtain a similar matching degree;
and obtaining a final matching degree based on the historical matching degree and the similarity matching degree, and performing matching processing of the final matching degree on the outlier sequence to obtain the abnormal behavior sequence.
Further, the processor 1001 may call the big data based grid operation management program stored in the memory 1005, and also perform the following operations:
judging whether the target date to be predicted in a short term is a vacation or not, if so, extracting a fluctuation sequence two days before the target date to be predicted in the short term and a synchronization sequence in the previous year, and outputting a prediction sequence;
further, the processor 1001 may call the big data based grid operation management program stored in the memory 1005, and also perform the following operations:
the step of judging whether the target date to be predicted in a short term is a holiday further comprises the following steps:
if the target date to be predicted in the short term is not a holiday, selecting a preset number of load sequences adjacent to the target date from the load sequence aggregation cluster, and judging whether the preset number of load sequences and the load sequences to be predicted in the short term are in the same time period;
if the preset number of load sequences and the load sequences to be predicted in a short term are in the same time period, intercepting the load sequences which are away from the target date by preset interval time from the selected load sequences to serve as reference sequences, and outputting the predicted sequences based on the reference sequences;
And if the preset number of load sequences and the load sequences to be predicted in the short term are not in the same time period, removing the load sequences to be predicted in the short term from the load sequence cluster.
Further, the processor 1001 may call the big data based grid operation management program stored in the memory 1005, and also perform the following operations:
the step of performing a pre-processing operation on the power load data comprises:
one or more of the following pre-treatment operations are employed,
performing data cleaning on the power load data by filling missing values, smoothing noise data and deleting outliers;
or, performing data integration on the power load data through deduplication processing;
or, performing data specification on the power load data through compression;
or, the power load data is subjected to data conversion processing through data discretization and hierarchical processing.
An embodiment of the present invention provides a power grid operation management method based on big data, and referring to fig. 2, fig. 2 is a schematic flow diagram of a first embodiment of the power grid operation management method based on big data according to the present invention.
In this embodiment, the power grid operation management method based on big data includes:
Step S10: acquiring power load data, and performing preprocessing operation on the power load data to clear abnormal data in the power load data.
In this embodiment, the collection of the power load data is realized by collecting through a collection network and a terminal device. The types of the acquisition methods of the power load data are various, such as a distribution and power utilization dispatching control center, a meter reading system, an intelligent control system of a street lamp, a monitoring security system of a power generation facility and the like. According to the different positions of the information of the power load data to be collected in the four links of transmission, distribution and utilization of the power grid system, the functional requirements of each collection system of the power load data are different. In the power generation process, according to power load data of different sources, such as hydroelectric power generation, wind power generation, thermal power generation, nuclear power generation and other new energy power generation modes, the construction of the whole management control system is different due to different power generation principles or modes; in the power transmission link, the difference of power transmission distance and lines causes different management modes, transformer substation construction numbers and the like; in a power distribution link, the power utilization requirements and power utilization levels of different areas in different time periods are greatly different, and a differentiated scheduling, management and monitoring system is required to be built; in the electricity utilization link, according to different types of electricity utilization categories of users, monitoring management in different degrees is required, such as the change condition of the electricity consumption of the users.
The intelligent management of the power system requires the timely feedback of power data, and the acquisition of power load data is very important. The power load data is easily interfered by noise, meanwhile, missing values and inconsistent data exist, and due to the fact that the data sources are different, the availability of the collected data is low, the difficulty of data mining is increased, and the mining effect is influenced. When the mining technology is used, in view of the fact that the data sources may be very different, and the formats, dimensions and the like of the data are different, preprocessing of load data is required in advance to improve the data availability, and then the data mining quality is improved. Data quality is generally affected by many factors, such as accuracy, integrity, consistency, timeliness, credibility and interpretability, and data that meets the requirements of an application is generally considered to be high-quality data.
Step S20: performing density clustering of fluctuation intervals on the power load data subjected to preprocessing operation to obtain an outlier sequence for abnormal detection and a load sequence aggregation cluster for prediction, establishing an abnormal behavior detection model according to the outlier sequence, and establishing a load prediction model according to the load sequence aggregation cluster.
The cluster analysis is a process of dividing a data object set into a plurality of groups or clusters with highly similar internal data through a certain calculation method, and the objects in different clusters have larger difference, namely the data are processed mainly by the principle of maximizing the similarity in the clusters and minimizing the similarity between the clusters. In the prior art, there are many clustering methods, and the basic clustering methods can be classified into the following methods according to different ideas used in the clustering method: the dividing method comprises the following steps: the partitioning method is performed by constructing a number of partitions smaller than the number of objects for a given data set, each partition representing a cluster containing at least one object. The hierarchical method comprises the following steps: the hierarchical method establishes different hierarchical divisions according to a data set, and is mainly classified into an aggregation method and a splitting method according to different methods. Grid-based methods: by translating the object space into a network structure of a limited number of cells, subsequent operations are then performed within the grid. Density-based methods: since the method of clustering using object distances can only find spherical clusters and it is difficult to find clusters of arbitrary shapes, a density-based clustering method has been developed.
In view of the fact that the power load data under normal conditions includes multidimensional data, clusters of various shapes may be formed during the Clustering process, in order to better detect abnormal electricity consumption behavior, in this embodiment, DBSCAN (Density Based Clustering method) is adopted, which is a Density Clustering method Based on high Density connected regions with Noise application, and determines the neighborhood radius of each object by setting a parameter greater than zero by a user, and using a parameter Min Pts to specify a Density threshold of a clustered cluster.
Step S30: and performing abnormal behavior detection on the outlier sequence by using the abnormal behavior detection model to obtain an abnormal behavior sequence, performing short-term load prediction on the load sequence aggregation cluster by using the load prediction model to obtain a prediction sequence, and managing the operation of the power grid based on the abnormal behavior sequence and the prediction sequence.
In the embodiment, the power abnormality includes not only a statistical condition that some dimension information in the power consumption data is obviously deviated from the dimension due to equipment failure, electricity stealing and the like, but also a condition that the power consumption data conforms to the statistical rule but violates the behavior rule of the user. In the former case, the abnormality can be detected well by setting threshold information for each dimension, and in the latter case, the electricity data is processed and analyzed by using a data mining method. Because the electricity utilization behaviors of the users have certain similarity in time and space, the current electricity utilization behaviors of the users are reflected to have similarity with the electricity utilization behaviors of the users in a certain historical period in time, and the electricity utilization behaviors of the users in the same region are reflected to have similar electricity utilization laws in space. Therefore, the abnormal behavior detection of the user can be realized by a data mining method according to the characteristics of the user electricity utilization behavior in the power system by utilizing the historical electricity utilization information. In the data mining technology, the density clustering DBSCAN method has the characteristics of capability of finding clustered clusters in any shapes and good anti-interference performance, and is relatively suitable for clustering of power loads.
In this embodiment, since the electrical load has its own characteristics in time and space, the variation trend of the load sequence can be obtained according to the load sequence by using the periodicity in time. Load sequence data is interval data, which is usually expressed as a geometric body in an n-dimensional space or as a cartesian product of spatial distribution. The interval data is represented by two methods, one is based on the upper and lower bounds of the value, and the other is the midpoint of the interval and the radius of the interval. When using time series for prediction, several consecutive load series can be considered as a whole, and the load data of every two consecutive days can be used as a load pattern. For the load mode of the day, the load mode of the day can be considered to influence the load mode of the day to a great extent in probability, and a plurality of similar load models are used for predicting the load change situation of the day to be predicted according to the load rules of the load models.
In the present embodiment, first, the power load data, the type of power abnormality, the demand for power load prediction, and the like are analyzed with respect to the characteristics of the power load; then preprocessing the power load data, clearing null values and noise data in the data, and improving data availability; then, clustering the fluctuation interval of the load sequence by using a density clustering technology so as to establish a corresponding abnormal behavior detection model and a corresponding load prediction model; and carrying out abnormal behavior detection and short-term load prediction on the power utilization load of the user by using the established model, and analyzing the detected result so as to provide information support for a management department.
In the embodiment, in view of the variety of sources of the load data, in order to improve the usability of the data, the method for analyzing the power data using time series is preprocessed; for the detection of abnormal behaviors, density clustering is carried out on fluctuation intervals of the power loads to obtain outliers, then matching processing is carried out by utilizing historical power utilization information of the user and power load characteristic curves of similar users of the user, and abnormal suspicion of the outliers is obtained through analysis; because the power dispatching can not be separated from the prediction of the power load, a density clustering prediction method based on a time sequence is provided according to the prediction demand of the power load and the demand. The method comprises the steps of representing power utilization modes by indexes such as starting time, fluctuation time span, starting power consumption, average growth rate and the like of a power utilization sequence, clustering the power utilization modes to obtain similar load cluster clusters, and performing short-term prediction on user loads by combining a regression method. During anomaly detection, relevance among similar users is rarely considered in the prior art, and when the power consumption behaviors of users of the same similar type are all changed, if the power consumption behaviors of a single user are only considered, misjudgment can be caused, and the anomaly detection effect is influenced. Therefore, a multi-user abnormal electricity utilization behavior detection method based on density clustering is provided. In the aspect of load prediction, in order to improve the prediction time precision, a method of clustering load fluctuation intervals is provided for short-term prediction of the electric load. Therefore, the detection of abnormal electricity utilization behaviors and the short-term prediction of power loads are realized, and the guarantee is provided for the safe and reliable operation of the power grid.
Optionally, the anomaly data includes noise data and missing value data,
the step of clearing abnormal data in the power load data comprises the following steps:
and cleaning the noise data and the missing value data according to a preset time sequence model.
In the present embodiment, the main parameters of the power load data include a user ID, a user category, user family condition information, a power usage category, a station number, a date and time, a voltage value, a current value, a power usage amount, power (active power and reactive power), a line loss value, and the like. Abnormal data in an electric power system can be classified into two cases, one is numerical errors including noise data and missing values; and the other is the transition of the whole state caused by the change of the operating environment, the change of the power mode is reflected in the condition, the research value is realized, and the further mining method can be used for further deep research after the influence of noise factors is eliminated.
In order to improve the effect of power load analysis, the reliability of load data should be ensured. Generally, the power load data is influenced by various factors, error data of problems such as noise data and missing values are generated, a corresponding model is built for the error data according to a time sequence,
Wherein:
which is indicative of the current load sequence,
indicating that the load sequence does not contain an abnormal value type, n indicates the number of abnormal values,
and
an influence factor indicating a load anomaly value,
the pulse function of the time instant is shown.
Optionally, the step of cleaning the noise data and the missing value data according to a preset time series model includes:
if the cleaned data object is the noise data, after the noise data is detected based on the front and back fluctuation relation of the load sequence and the distance between the loads, adopting the change data of the current load relative to the time sequence of the previous preset time period to repair;
if the cleaned data object is the missing value data, acquiring load data of the same time point of adjacent preset time periods before and after the missing value in the load sequence, calculating according to the load data of the same time point to obtain a load average value, obtaining a load variation based on the load variation rate of the next time period relative to the previous time period, and adding the load variation to fill the missing value data on the basis of the load average value.
In this embodiment, when load data is preprocessed, corresponding processing methods are respectively designed according to different exception types.
For noise data, the noise data shows a fluctuation range deviating from the normal curve or even a burr appears on the whole load curve. The detection of such anomalies can be based on the relationship of the fluctuations before and after the load sequence and the distance between the loads. That is, the distance of the k loads closest to the central load point s in the load sequence is taken as the density region of the sequence, and the region radius of the point with higher density is smaller, and the larger of the distance d from the load point to the central point and the density region radius r is taken as the reachable distance of the central point s. The abnormal degree of the load can be embodied by the relative distance from the load point to the central point, and the abnormal threshold value is set, so that the load point with abnormal data can be obtained by judgment.
For the missing value, since the power load data generally has a fluctuation cycle characteristic, loads at the same time points on two adjacent days before and after the missing value and an average value of loads at two time points before and after the missing value are calculated according to the characteristic, and the missing value is filled by adding a load change amount to the load average value in combination with a load change rate method on the day before and after.
Optionally, the step of performing density clustering on fluctuation intervals on the power load data after the preprocessing operation to establish an abnormal behavior detection model and a load prediction model includes:
Performing density clustering of fluctuation intervals on the power load data subjected to the preprocessing operation to obtain an outlier sequence for abnormal detection and a load sequence aggregation cluster for prediction;
and establishing an abnormal behavior detection model according to the outlier sequence, and establishing a load prediction model according to the load sequence cluster.
In this embodiment, the power load data has its own regularity and similarity in time and space, the time is represented as that the power load condition of the current time period has a great similarity to some historical time periods, and the space is represented as that the current user has a similar power consumption behavior to the same type of users in the same region. Therefore, the cluster analysis is carried out by using the power load data, and load aggregation clusters with similar power consumption behaviors of the same type and outlier power consumption data with less frequency of occurrence can be obtained. The detection of the abnormal behavior of the power utilization can be realized by further analyzing the outlier data according to the clustering result, and the load prediction result in a super short term can be obtained by utilizing the data in the clustering cluster in combination with the regression idea and the time series analysis method, so that the purpose of load prediction is realized. And performing density clustering of fluctuation intervals on the preprocessed power load data to obtain an outlier load sequence for anomaly detection and a load sequence clustering cluster for prediction, and further performing analysis processing.
Optionally, the step of performing abnormal behavior detection on the outlier sequence by using the abnormal behavior detection model to obtain an abnormal behavior sequence includes:
reading user information to obtain a load sequence of a user in a preset reading period, performing cluster analysis on the load sequence to obtain a similar category group, and judging whether an outlier sequence exists in the load sequence based on the similar category group;
if the outlier sequence exists, calculating the outlier sequence by using a preset historical data model to obtain a historical matching degree, and calculating the outlier sequence by using a preset similar user model to obtain a similar matching degree;
and obtaining a final matching degree based on the historical matching degree and the similarity matching degree, and performing matching processing of the final matching degree on the outlier sequence to obtain the abnormal behavior sequence.
In this embodiment, in the case of normal power consumption, the voltage, current, power factor, power consumption, and line loss of the user are kept relatively stable, and if the variation range of a certain attribute value is too large, it means that an abnormality may occur. The voltage, current, power factor, power consumption and line loss may all cause power abnormality. The power utilization abnormity of the user can be caused by various factors such as line faults, disconnection, power supply faults, meter faults, external interference, data transmission errors, power theft and the like. The power abnormality generally has persistence, and many times, more than one kind of abnormality occurs, and multiple kinds of abnormalities are likely to occur simultaneously. As shown in fig. 3, many factors need to be considered in the division of the electricity consumers, and the electricity information of one consumer generally includes information such as electricity type, voltage level, power factor, daily electricity amount, daily average load, daily maximum load, daily minimum load, daily peak-to-valley difference, peak-to-total ratio, average-to-total ratio, valley-to-total ratio, load rate, and load curve. In addition, environmental factors such as a transformer area, weather and temperature are considered among different users.
When the historical electricity utilization data of the users are used for carrying out abnormity detection, in view of the fact that the external environment factors such as the geography of the same region, the climate condition and the like are basically the same, the users with the same electricity utilization type have similar electricity utilization behaviors, the electricity utilization behaviors of the users per se have similarity in the time dimension, namely the historical electricity utilization load data presents certain clustering characteristics, and therefore the electricity utilization models of the users can be respectively established. The power utilization behaviors of the user in each week in a month and in the same time period every day in the same week have certain correlation, so that the power utilization load sequence of the abnormal sequence of the user in the same time period every day in the week is extracted, and the power utilization sequence of the same time period in the same week in the month is used for establishing a historical power utilization model of the user. And selecting load sequence representation of the outlier sequence in the same time period by using the electricity utilization model of the users with similar types. The similar users are selected by comprehensively considering factors such as the power utilization environment of the users, the characteristics of the users and the like, the power utilization behavior characteristics of a certain class of users are reflected, and for the judged similar users, the power utilization behavior characteristics of a certain user can represent the power utilization behaviors of the class of users.
And for the abnormal electricity utilization behavior of the user, clustering the user information in the database to obtain a similar user category group. Then, the historical load data of the user to be investigated is quantized and clustered, and if the clustering result shows an outlier load sequence, the sequence is added into an outlier sequence library. And constructing a historical electricity utilization data model and a similar user electricity utilization model of the user according to the outlier sequence, matching the outlier load sequence, determining the outlier suspicion degree according to a matching result, and finally judging whether the electricity utilization behavior is abnormal or not.
In this embodiment, a data mining method is used to detect abnormal electricity consumption behaviors of users, a density clustering-based detection algorithm for abnormal electricity consumption behaviors is provided, the algorithm utilizes historical data clustering to obtain different user type groups and negative sequence cluster, and the cluster load sequence is matched with the historical electricity consumption model of the user and the electricity consumption model of a similar user, so that the detection of the electricity consumption behaviors of multiple users is realized.
Optionally, the step of performing short-term prediction of load on the load sequence aggregation cluster by using the load prediction model to obtain a prediction sequence includes:
judging whether the target date to be predicted in a short term is a vacation or not, if so, extracting a fluctuation sequence and a previous year synchronization sequence two days before the target date to be predicted in the short term, and outputting a prediction sequence based on the fluctuation sequence and the previous year synchronization sequence.
In this embodiment, in general, the closer the load prediction is to the time to be predicted, the more similar the load sequence in the cluster is to the load sequence to be predicted, i.e., the more recent or recent characteristics in the load prediction. When a user normally uses electricity, the electricity utilization behavior in the previous period has certain correlation with the electricity utilization behavior in the next period, and according to the thought of time sequence, the electricity utilization correlation characteristics of two same time periods before and after the previous day can be utilized to predict the next electricity utilization situation according to the electricity utilization fluctuation situation of the previous sequence to be predicted. The method is difficult to ensure that the electricity utilization behaviors in the time periods before and after the day are similar to the electricity utilization behaviors to be predicted, and has certain limitation.
When prediction is carried out, a fluctuation interval of a moment before a time point to be predicted is selected, and the load in the aggregation cluster where the power utilization mode is located is used for prediction. And selecting 30 load sequences in the same time period closest to the time to be predicted from the aggregation cluster for the sequences to be predicted in the non-holidays, and eliminating the load sequences which have the time distance exceeding one week and are not the same week number in one month in the sequences, namely, retaining the sequences in the same week days in the week and the same week days in the month as the days to be predicted. The payload sequence for the next time segment of the reserved sequence is then extracted as the reference sequence. And selecting load sequences within two days before the date to be predicted and within the same date and two days before the date to be predicted in the last year as the date to be predicted from the aggregation cluster, and extracting the load sequence in the next time period as a reference sequence according to the selected sequences.
Optionally, the step of determining whether the target date to be short-term predicted is after the vacation further comprises:
if the target date to be predicted in the short term is not a holiday, selecting a preset number of load sequences adjacent to the target date from the load sequence aggregation cluster, and judging whether the preset number of load sequences and the load sequences to be predicted in the short term are in the same time period;
If the preset number of load sequences and the load sequences to be predicted in the short term are in the same time period, intercepting the load sequences which are away from the target date by preset interval time from the selected load sequences as reference sequences, and outputting the predicted sequences based on the reference sequences;
and if the preset number of load sequences and the load sequences to be predicted in the short term are not in the same time period, removing the load sequences to be predicted in the short term from the load sequence cluster.
In this embodiment, as shown in fig. 4, when load prediction is performed, historical electrical load information of a user is read from a database, load data is preprocessed, load sections are divided, and indexes of the sections are calculated. And then, according to the clustering result of the last fluctuation sequence of the time to be predicted, judging whether the date to be predicted is in a vacation period, and then selecting a reference sequence from the clustering cluster. For the date to be predicted is not a holiday, selecting 30 sequences nearest to the date to be predicted from the cluster, selecting the sequences with the same days of the week as the reference sequences from the sequences, and constructing a prediction model; and when the day to be predicted is a holiday, extracting a next fluctuation sequence of the two days before the day to be predicted and a load sequence of the last three days in the same period of the last year and the same time period to be predicted to construct a prediction model, and finally predicting the load to obtain a load prediction result.
In this embodiment, the fluctuation intervals of the sequence are divided according to the periodicity and regularity of the historical power consumption load and the variation trend of the load, and the features of the load intervals extracted from the fluctuation intervals are expressed as power consumption patterns and subjected to density clustering. And selecting similar load sequences from the cluster clusters according to the clustering result, determining a reference sequence, and establishing a load prediction model by combining a time sequence and a regression idea to predict the load, so that the ultra-short-term load prediction of the load is realized, and a prediction method based on the power utilization mode is provided for a power management department.
Optionally, the step of performing a preprocessing operation on the power load data comprises:
one or more of the following pre-treatment operations are employed,
performing data cleaning on the power load data by filling missing values, smoothing noise data and deleting outliers;
or, performing data integration on the power load data through deduplication processing;
or, performing data specification on the power load data through compression;
or, the power load data is subjected to data conversion processing through data discretization and hierarchical processing.
In this embodiment, there are many methods for data preprocessing, and the main steps of data preprocessing can be divided into data cleaning, integration, specification and transformation. Specifically, data cleansing is accomplished by filling in missing values, smoothing out noisy data, identifying or deleting outliers, and resolving inconsistencies. Data integration is the process of dealing with the problem that data from different systems, databases or files have different names in different databases when representing the same concept, resulting in data inconsistency and redundancy. Data conventions are mainly used to simplify data sets, including using data encoding schemes such that the original data is given simplified, compressed dimension conventions; the two forms of numerical reduction representing the original data are replaced in a smaller way by methods of parametric models represented by regression and log-linear models and non-parametric models represented by clustering, decimation and data aggregation. Data transformation is the alteration of the data form to achieve normalization, data discretization, and concept layering for powerful data mining using discretization and concept layering at multiple abstraction levels.
In addition, an embodiment of the present invention further provides a device for managing power grid operation based on big data, where the device includes:
the preprocessing module is used for acquiring power load data, preprocessing the power load data and clearing abnormal data in the power load data;
the density clustering module is used for performing density clustering of fluctuation intervals on the power load data after the preprocessing operation, and establishing an abnormal behavior detection model and a load prediction model;
and the detection prediction module is used for performing abnormal behavior detection on the outlier sequence by using the abnormal behavior detection model to obtain an abnormal behavior sequence, performing short-term prediction on the load of the load sequence aggregation cluster by using the load prediction model to obtain a prediction sequence, and managing the operation of the power grid based on the abnormal behavior sequence and the prediction sequence.
In addition, an embodiment of the present invention further provides a big data-based power grid operation management device, where the big data-based power grid operation management device includes: the system comprises a memory, a processor and a big data-based power grid operation management program stored on the memory and capable of running on the processor, wherein the big data-based power grid operation management program realizes the steps of the big data-based power grid operation management method when being executed by the processor.
In addition, an embodiment of the present invention further provides a computer-readable storage medium, where a big data-based power grid operation management program is stored on the computer-readable storage medium, and when executed by a processor, the big data-based power grid operation management program implements the steps of the big data-based power grid operation management method described above.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are only for description, and do not represent the advantages and disadvantages of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) as described above and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. The power grid operation management method based on big data is characterized by comprising the following steps of:
acquiring power load data, and cleaning abnormal data in the power load data by performing preprocessing operation on the power load data;
performing density clustering of fluctuation intervals on the power load data subjected to preprocessing operation to obtain an outlier sequence for abnormal detection and a load sequence aggregation cluster for prediction, establishing an abnormal behavior detection model according to the outlier sequence, and establishing a load prediction model according to the load sequence aggregation cluster;
and performing abnormal behavior detection on the outlier sequence by using the abnormal behavior detection model to obtain an abnormal behavior sequence, performing short-term load prediction on the load sequence aggregation cluster by using the load prediction model to obtain a prediction sequence, and managing the operation of the power grid based on the abnormal behavior sequence and the prediction sequence.
2. The big-data based power grid operation management method according to claim 1, wherein the abnormal data includes noise data and missing value data,
the step of clearing abnormal data in the power load data comprises the following steps:
and cleaning the noise data and the missing value data according to a preset time sequence model.
3. The big data-based power grid operation management method according to claim 2, wherein the step of cleaning the noise data and the missing value data according to a preset time series model comprises:
if the cleaned data object is the noise data, after the noise data is detected based on the front and back fluctuation relation of the load sequence and the distance between the loads, adopting the change data of the current load relative to the time sequence of the previous preset time period to repair;
if the cleaned data object is the missing value data, acquiring load data of the same time point of adjacent preset time periods before and after the missing value in the load sequence, calculating according to the load data of the same time point to obtain a load average value, obtaining a load variation based on the load variation rate of the next time period relative to the previous time period, and adding the load variation to fill the missing value data on the basis of the load average value.
4. The big data-based power grid operation management method according to claim 1, wherein the step of performing abnormal behavior detection on the outlier sequence by using the abnormal behavior detection model to obtain the abnormal behavior sequence comprises:
reading user information to obtain a load sequence of a user in a preset reading period, performing cluster analysis on the load sequence to obtain a similar category group, and judging whether an outlier sequence exists in the load sequence based on the similar category group;
if the outlier sequence exists, calculating the outlier sequence by using a preset historical data model to obtain a historical matching degree, and calculating the outlier sequence by using a preset similar user model to obtain a similar matching degree;
and obtaining a final matching degree based on the historical matching degree and the similarity matching degree, and performing matching processing of the final matching degree on the outlier sequence to obtain the abnormal behavior sequence.
5. The big data based power grid operation management method according to claim 1, wherein the step of performing short-term load prediction on the load sequence aggregation cluster by using the load prediction model to obtain a prediction sequence comprises:
Judging whether the target date to be predicted in a short term is a vacation or not, if so, extracting a fluctuation sequence and a previous year synchronization sequence two days before the target date to be predicted in the short term, and outputting a prediction sequence based on the fluctuation sequence and the previous year synchronization sequence.
6. The big data based grid operation management method according to claim 5, wherein the step of determining whether the target date to be short-term predicted is a holiday further comprises:
if the target date to be predicted in the short term is not a holiday, selecting a preset number of load sequences adjacent to the target date from the load sequence aggregation cluster, and judging whether the preset number of load sequences and the load sequences to be predicted in the short term are in the same time period;
if the preset number of load sequences and the load sequences to be predicted in a short term are in the same time period, intercepting the load sequences which are away from the target date by preset interval time from the selected load sequences to serve as reference sequences, and outputting the predicted sequences based on the reference sequences;
and if the preset number of load sequences and the load sequences to be predicted in the short term are not in the same time period, removing the load sequences to be predicted in the short term from the load sequence aggregation cluster.
7. The big data based power grid operation management method of claim 1, wherein the step of pre-processing the power load data by the pre-processing operation comprises:
one or more of the following pre-treatment operations are employed,
performing data cleaning on the power load data by filling missing values, smoothing noise data and deleting outliers;
or, performing data integration on the power load data through deduplication processing;
or, performing data specification on the power load data through compression;
or, the power load data is subjected to data conversion processing through data discretization and hierarchical processing.
8. A big data based grid operation management apparatus, the apparatus comprising:
the preprocessing module is used for acquiring power load data, preprocessing the power load data and clearing abnormal data in the power load data;
the density clustering module is used for performing density clustering of fluctuation intervals on the power load data after the preprocessing operation, and establishing an abnormal behavior detection model and a load prediction model;
and the detection prediction module is used for performing abnormal behavior detection on the outlier sequence by using the abnormal behavior detection model to obtain an abnormal behavior sequence, performing short-term prediction on the load of the load sequence aggregation cluster by using the load prediction model to obtain a prediction sequence, and managing the operation of the power grid based on the abnormal behavior sequence and the prediction sequence.
9. A big data based power grid operation management device, comprising: a memory, a processor, and a big-data based grid operation management program stored on the memory and executable on the processor, the big-data based grid operation management program configured to implement the steps of the big-data based grid operation management method according to any one of claims 1 to 7.
10. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a big data based grid operation management program, which when executed by a processor implements the steps of the big data based grid operation management method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210196800.XA CN114676883A (en) | 2022-03-02 | 2022-03-02 | Power grid operation management method, device and equipment based on big data and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210196800.XA CN114676883A (en) | 2022-03-02 | 2022-03-02 | Power grid operation management method, device and equipment based on big data and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114676883A true CN114676883A (en) | 2022-06-28 |
Family
ID=82072293
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210196800.XA Pending CN114676883A (en) | 2022-03-02 | 2022-03-02 | Power grid operation management method, device and equipment based on big data and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114676883A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114977176A (en) * | 2022-07-19 | 2022-08-30 | 深圳江行联加智能科技有限公司 | Power load decomposition method, device, equipment and storage medium |
| CN115510302A (en) * | 2022-11-16 | 2022-12-23 | 西北工业大学 | Intelligent factory data classification method based on big data statistics |
| CN115982602A (en) * | 2023-03-20 | 2023-04-18 | 济宁众达利电气设备有限公司 | Photovoltaic transformer electrical fault detection method |
| CN116243097A (en) * | 2023-05-11 | 2023-06-09 | 新风光电子科技股份有限公司 | Electric energy quality detection method based on big data |
| CN116307944A (en) * | 2023-05-18 | 2023-06-23 | 山东欣瑞电气有限公司 | Distribution box remote monitoring system based on artificial intelligence and Internet of things |
| CN116628529A (en) * | 2023-07-21 | 2023-08-22 | 山东科华电力技术有限公司 | Data anomaly detection method for intelligent load control system at user side |
| CN117220417A (en) * | 2023-11-07 | 2023-12-12 | 国网山西省电力公司信息通信分公司 | Dynamic monitoring method and system for consumer-side electrical load |
| CN117405971A (en) * | 2023-10-09 | 2024-01-16 | 国网河南电力公司营销服务中心 | Power acquisition digitization method based on flow automation |
| CN117454201A (en) * | 2023-12-26 | 2024-01-26 | 无锡市锡容电力电器有限公司 | Method and system for detecting abnormal operation state of smart power grid |
| CN117977717A (en) * | 2024-04-01 | 2024-05-03 | 国网黑龙江省电力有限公司佳木斯供电公司 | Cold region wind-solar-thermal energy storage comprehensive energy collaborative management method and system |
| WO2024119919A1 (en) * | 2022-12-08 | 2024-06-13 | 云南电网有限责任公司信息中心 | Power-grid operation state monitoring method and system based on multi-source distribution-network monitoring data |
| CN118449275A (en) * | 2024-05-06 | 2024-08-06 | 怀化建南机器厂有限公司 | Intelligent self-adjusting electric power concentrator |
-
2022
- 2022-03-02 CN CN202210196800.XA patent/CN114676883A/en active Pending
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114977176A (en) * | 2022-07-19 | 2022-08-30 | 深圳江行联加智能科技有限公司 | Power load decomposition method, device, equipment and storage medium |
| CN115510302A (en) * | 2022-11-16 | 2022-12-23 | 西北工业大学 | Intelligent factory data classification method based on big data statistics |
| WO2024119919A1 (en) * | 2022-12-08 | 2024-06-13 | 云南电网有限责任公司信息中心 | Power-grid operation state monitoring method and system based on multi-source distribution-network monitoring data |
| CN115982602A (en) * | 2023-03-20 | 2023-04-18 | 济宁众达利电气设备有限公司 | Photovoltaic transformer electrical fault detection method |
| CN116243097A (en) * | 2023-05-11 | 2023-06-09 | 新风光电子科技股份有限公司 | Electric energy quality detection method based on big data |
| CN116243097B (en) * | 2023-05-11 | 2023-08-15 | 新风光电子科技股份有限公司 | Electric energy quality detection method based on big data |
| CN116307944A (en) * | 2023-05-18 | 2023-06-23 | 山东欣瑞电气有限公司 | Distribution box remote monitoring system based on artificial intelligence and Internet of things |
| CN116307944B (en) * | 2023-05-18 | 2023-08-04 | 山东欣瑞电气有限公司 | Distribution box remote monitoring system based on artificial intelligence and Internet of things |
| CN116628529B (en) * | 2023-07-21 | 2023-10-20 | 山东科华电力技术有限公司 | Data anomaly detection method for intelligent load control system at user side |
| CN116628529A (en) * | 2023-07-21 | 2023-08-22 | 山东科华电力技术有限公司 | Data anomaly detection method for intelligent load control system at user side |
| CN117405971A (en) * | 2023-10-09 | 2024-01-16 | 国网河南电力公司营销服务中心 | Power acquisition digitization method based on flow automation |
| CN117220417A (en) * | 2023-11-07 | 2023-12-12 | 国网山西省电力公司信息通信分公司 | Dynamic monitoring method and system for consumer-side electrical load |
| CN117220417B (en) * | 2023-11-07 | 2024-02-09 | 国网山西省电力公司信息通信分公司 | Dynamic monitoring method and system for consumer-side electrical load |
| CN117454201A (en) * | 2023-12-26 | 2024-01-26 | 无锡市锡容电力电器有限公司 | Method and system for detecting abnormal operation state of smart power grid |
| CN117454201B (en) * | 2023-12-26 | 2024-03-08 | 无锡市锡容电力电器有限公司 | Method and system for detecting abnormal operation state of smart power grid |
| CN117977717A (en) * | 2024-04-01 | 2024-05-03 | 国网黑龙江省电力有限公司佳木斯供电公司 | Cold region wind-solar-thermal energy storage comprehensive energy collaborative management method and system |
| CN117977717B (en) * | 2024-04-01 | 2024-06-11 | 国网黑龙江省电力有限公司佳木斯供电公司 | Cold region wind-solar-thermal energy storage comprehensive energy collaborative management method and system |
| CN118449275A (en) * | 2024-05-06 | 2024-08-06 | 怀化建南机器厂有限公司 | Intelligent self-adjusting electric power concentrator |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114676883A (en) | Power grid operation management method, device and equipment based on big data and storage medium | |
| CN110610280B (en) | Short-term prediction method, model, device and system for power load | |
| CN113962364A (en) | Multi-factor power load prediction method based on deep learning | |
| Zhao et al. | User electricity behavior analysis based on K-means plus clustering algorithm | |
| CN108446396B (en) | Power data processing method based on improved CIM model | |
| Khan et al. | Segmentation of factories on electricity consumption behaviors using load profile data | |
| CN115564071A (en) | Method and system for generating data labels of power Internet of things equipment | |
| Mohamed et al. | A review on big data management and decision-making in smart grid | |
| CN107862459B (en) | Metering equipment state evaluation method and system based on big data | |
| CN115730749B (en) | Power dispatching risk early warning method and device based on fusion power data | |
| Zhang et al. | Load forecasting method based on improved deep learning in cloud computing environment | |
| Zhang et al. | The power big data-based energy analysis for intelligent community in smart grid | |
| CN110543499A (en) | Anti-electricity-stealing location tracking equipment based on multi-source data fusion | |
| CN118353011B (en) | Method and system for detecting panoramic dynamic topology anomaly of power distribution network based on graph calculation | |
| CN117955245A (en) | Method and device for determining running state of power grid, storage medium and electronic equipment | |
| CN112835949A (en) | Load prediction method | |
| CN114372835B (en) | Comprehensive energy service potential customer identification method, system and computer equipment | |
| Zhang et al. | Research on scheduling strategy of heterogeneous flexible load clusters with strong association consider dual-scale similarities | |
| CN114168662A (en) | Power distribution network problem combing and analyzing method and system based on multiple data sources | |
| Dhupia et al. | A review: Big data analytics in smart grid management | |
| Sun et al. | Cloud-based data analysis of user side in smart grid | |
| Wang et al. | Application and Analysis of Big Data Technology in Smart Grid | |
| Yanhui et al. | Forecasting hourly electricity demand with nonparametric functional data analysis | |
| Dong et al. | Image-based with peak load ensemble prediction system for demand response in smart grid | |
| Yang et al. | Power user portrait model based on random forest |
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 |