CN117040106A - Boosting station active early warning method and system based on AI time sequence twinning technology - Google Patents

Boosting station active early warning method and system based on AI time sequence twinning technology Download PDF

Info

Publication number
CN117040106A
CN117040106A CN202310760831.8A CN202310760831A CN117040106A CN 117040106 A CN117040106 A CN 117040106A CN 202310760831 A CN202310760831 A CN 202310760831A CN 117040106 A CN117040106 A CN 117040106A
Authority
CN
China
Prior art keywords
data
time sequence
digital twin
booster station
monitoring data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310760831.8A
Other languages
Chinese (zh)
Inventor
刘晓冬
梁清鹤
魏江哲
马建宝
胡玉
张磊
梁志明
王超
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guoneng Ningdong New Energy Co ltd
Original Assignee
Guoneng Ningdong New Energy Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guoneng Ningdong New Energy Co ltd filed Critical Guoneng Ningdong New Energy Co ltd
Priority to CN202310760831.8A priority Critical patent/CN117040106A/en
Publication of CN117040106A publication Critical patent/CN117040106A/en
Pending legal-status Critical Current

Links

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
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/067—Enterprise or organisation modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • 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
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
    • H02J13/00032—Systems characterised by the controlled or operated power network elements or equipment, the power network elements or equipment not otherwise provided for
    • H02J13/00036—Systems characterised by the controlled or operated power network elements or equipment, the power network elements or equipment not otherwise provided for the elements or equipment being or involving switches, relays or circuit breakers
    • H02J13/0004—Systems characterised by the controlled or operated power network elements or equipment, the power network elements or equipment not otherwise provided for the elements or equipment being or involving switches, relays or circuit breakers involved in a protection system
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/12—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/24—Arrangements for preventing or reducing oscillations of power in networks
    • H02J3/241—The oscillation concerning frequency
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2123/00—Data types
    • G06F2123/02—Data types in the time domain, e.g. time-series data
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Power Engineering (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Tourism & Hospitality (AREA)
  • Software Systems (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Evolutionary Biology (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Quality & Reliability (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Game Theory and Decision Science (AREA)
  • Water Supply & Treatment (AREA)
  • Primary Health Care (AREA)
  • Educational Administration (AREA)
  • Medical Informatics (AREA)
  • Operations Research (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Development Economics (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The application relates to the technical field of booster station safety management, and discloses a booster station active early warning method and system based on an AI time sequence twinning technology, wherein the technical scheme is characterized by comprising the following steps: acquiring historical monitoring data of the booster station, and processing the historical monitoring data to obtain a monitoring data time sequence; dividing the monitoring data time sequence into a monitoring data time sequence sample set and a monitoring data time sequence test set; substituting the monitoring data time sequence sample set into a digital twin body data model, and training to obtain a primary training digital twin body data model; testing the initial training digital twin body data model by adopting a monitoring data time sequence test set to obtain a mature digital twin body data model; and constructing the mature digital twin database model into an operation framework of the AI digital control platform, accessing online monitoring data of the booster station to calculate to obtain simulation result data, and starting a corresponding early warning scheme if the simulation result data is matched with the early warning data template.

Description

Boosting station active early warning method and system based on AI time sequence twinning technology
Technical Field
The application relates to the technical field of safety management of booster stations, in particular to a booster station active early warning method and system based on an AI time sequence twinning technology.
Background
A booster station refers to a facility for raising or lowering the transmission voltage to a desired voltage level. Typically, a booster station is primarily responsible for boosting electrical energy for long distance transmission, as well as coordinating voltage and frequency between grids. The input end of the booster station is connected with the power transmission line, and the output end of the booster station is connected with a load or other power grids.
The booster station generally comprises a transformer, a circuit breaker, a disconnecting switch, a capacitor, a reactor, an automatic voltage regulating device, a protection device and the like. The transformer is one of the most important devices of the booster station, and is used for boosting or reducing the power transmission electric energy. And the isolating switch and the circuit breaker are used for controlling and protecting the transmission line and the equipment. The capacitors and reactors are used to control the grid voltage and reduce the grid harmonics. The automatic voltage regulating device is used for monitoring and controlling the voltage of the power grid.
The booster station is typically located at the end of the power system, remote from the electrical load center. They are typically built in open areas to accommodate large equipment and transmission lines, while also facilitating maintenance and repair. The booster station is mainly used for boosting the voltage level so as to realize long-distance transmission and coordinate the voltage and frequency between the power grids.
Along with the development of scientific technology, the requirements on the power supply reliability are higher and higher, the booster station operates in a high-voltage and high-current state for a long time, faults are easy to generate, the faults are mainly concentrated at the connection part of the line, and due to severe weather changes, storm wind, storm rain, snow storm and lightning, the thermal effect of the power transmission line is often aggravated, the ageing of metal materials and the reduction of mechanical properties are caused, partial discharge is generated, the burning loss, fusion welding or fracture damage of connecting pieces are caused, and the safe operation of the high-voltage power transmission line is directly influenced.
Therefore, an active early warning method aiming at the safety maintenance of the booster station is needed to be provided, and the safety of the power grid is ensured.
Disclosure of Invention
The application aims to provide a booster station active early warning method and system based on an AI time sequence twinning technology, wherein the booster station state is subjected to data discrimination acquisition to generate a digital twinning body data model constructed according to the booster station according to training set training, so that the booster station active early warning method and system have more accurate simulation result data, and further have more accurate early warning effects in the follow-up intelligent active early warning.
The technical aim of the application is realized by the following technical scheme: a booster station active early warning method based on an AI time sequence twinning technology comprises the following steps:
acquiring historical monitoring data of the booster station, and processing the historical monitoring data to obtain a plurality of monitoring data time sequences with set sample time lengths;
dividing a plurality of monitoring data time sequences into a monitoring data time sequence sample set and a monitoring data time sequence test set;
constructing a digital twin body data model of the transformer substation through basic data in the booster station, substituting a monitoring data time sequence sample set into the digital twin body data model, and training to obtain a primary training digital twin body data model;
testing the initial training digital twin body data model by adopting a monitoring data time sequence test set, and performing parameter debugging in the test until the test passes, so as to obtain a mature digital twin body data model;
after the mature digital twin body data model is obtained, an AI digital control platform is constructed, and the mature digital twin body database model is constructed into an operation framework of the AI digital control platform to obtain the AI digital control platform capable of operating the mature digital twin body database model;
after the AI digital control platform is obtained, the online monitoring data of the booster station is accessed, the online monitoring data is input into the AI digital control platform to run the mature digital twin database model in real time for calculation, the simulation result data is obtained, and if the simulation result data is matched with a preset early warning data template, the corresponding early warning scheme is actively started.
As a preferred technical solution of the present application, after the history monitoring data of the booster station is obtained, the processing of the history monitoring data includes the following steps: and carrying out structuring treatment on the obtained historical monitoring data, and extracting multidimensional time series data of each monitoring item.
When the historical monitoring data of the booster station is obtained, the historical fault data of the booster station in a fault state and the historical normal data of the booster station in a normal running state are respectively obtained; screening and dividing the historical fault data and the historical normal data in a standard sampling manner respectively to obtain a plurality of continuous fault data time sequences and normal data time sequences with set sample time lengths;
selecting a fault data time sequence sample set and a fault data time sequence test set from the obtained fault data time sequences according to a set selection proportion; selecting a normal data time sequence sample set and a normal data time sequence test set from the obtained normal data time sequences according to a set selection proportion;
and substituting the fault data time series sample set and the normal data time series sample set into training respectively when training the digital twin volume data model.
According to the application, when the historical fault data and the historical normal data are screened and divided in a standard sampling manner, the detailed state of the power grid corresponding to the time of the historical fault data and the time of the historical normal data is obtained, and when the fault data time sequence and the normal data time sequence are obtained through division, the detailed state of the power grid corresponding to the time in the fault data time sequence and the normal data time sequence is associated.
As a preferable technical scheme of the application, the ratio of the historical data time series sample set to the historical data time series test set in the dividing process is 4:1.
As a preferable technical solution of the present application, the basic data in the booster station includes: attribute data, topological connection relation, equipment operation data and environment data of physical nodes where all equipment in the booster station are located.
As a preferable technical scheme of the application, after the AI digital control platform is put into use, the evaluation data of the simulation result data of a user is obtained at regular time, and the parameters of the mature digital twin body data model are optimized according to the evaluation data.
As a preferable technical scheme of the application, after the optimized parameters are calculated by the mature digital twin data model, the method enters a human approval process to obtain a human approval result, and the parameters used by the mature digital twin data model are determined according to the human approval result.
A booster station active early warning system based on AI time sequence twinning technology comprises:
the historical monitoring data processing module is used for acquiring historical monitoring data of the booster station and processing the historical monitoring data to obtain a plurality of monitoring data time sequences with set sample time lengths;
the training set dividing module divides the plurality of monitoring data time sequences into a monitoring data time sequence sample set and a monitoring data time sequence test set;
the digital twin body data model training module is used for constructing a digital twin body data model of the transformer substation through basic data in a booster station, substituting a monitoring data time sequence sample set into the digital twin body data model, and training to obtain a primary training digital twin body data model;
the digital twin body data model test module adopts a monitoring data time sequence test set to test the initial training digital twin body data model, and performs parameter debugging in the test until the test passes, so as to obtain a mature digital twin body data model;
the AI digital control platform module is used for constructing an AI digital control platform after the mature digital twin body data model is obtained, and constructing the mature digital twin body database model into an operation framework of the AI digital control platform to obtain the AI digital control platform capable of operating the mature digital twin body database model;
and the active early warning module is used for accessing the online monitoring data of the booster station after the AI digital control platform is obtained, inputting the online monitoring data into the AI digital control platform to run the mature digital twin database model in real time for calculation to obtain simulation result data, and actively starting a corresponding early warning scheme if the simulation result data is matched with a preset early warning data template.
In summary, the application has the following beneficial effects: when the historical monitoring data are acquired, data are acquired in a distinguishing mode according to the state of the booster station, the historical fault data of the booster station in a fault state and the historical normal data of the booster station in a normal operation state are respectively acquired, and meanwhile, the historical monitoring data in the two states are respectively processed to obtain time sequence training sets in the two states. The digital twin body data model is trained by utilizing the time sequence training sets of the two states respectively to obtain the parameters of the digital twin body data model attached to the booster station to be served currently, and the digital twin body data model is built according to the booster station to be served, so that more accurate simulation result data can be provided during actual service, and further more accurate early warning effect is provided during follow-up intelligent active early warning.
Drawings
Fig. 1 is a flow chart of the method of the present application.
Detailed Description
The present application will be described in further detail with reference to the accompanying drawings.
As shown in fig. 1, the application provides a booster station active early warning method and a booster station active early warning system based on an AI time sequence twinning technology, wherein the method is executed by adopting modules in the system, and specifically, the booster station active early warning method based on the AI time sequence twinning technology comprises the following steps:
s1, acquiring historical monitoring data of a booster station, and processing the historical monitoring data to obtain a plurality of monitoring data time sequences with set sample time lengths;
specifically, after the historical monitoring data of the booster station are obtained, the obtained historical monitoring data are processed, the structural processing is carried out according to the data composition of the historical monitoring data, and the multidimensional time series data of each monitoring item are extracted so as to facilitate the processing training of a subsequent digital twin body data model.
When the historical monitoring data of the booster station is acquired, the historical fault data of the booster station in a fault state and the historical normal data of the booster station in a normal running state are required to be acquired respectively; screening and dividing the historical fault data and the historical normal data in a standard sampling manner respectively to obtain a plurality of continuous fault data time sequences and normal data time sequences with set sample time lengths;
the screening and dividing of the standard samples of the historical fault data and the historical normal data are that the missing data in the historical fault data and the historical normal data are filled with average data, so that the data have integrity, the data are screened from the historical fault data and the historical normal data through the format of a pre-established standard sample, the data which do not accord with the standard sample format are removed, the data are divided from the data which accord with the standard sample format, the dividing basis is the set sample time length, and finally the fault data time sequence and the normal data time sequence are obtained.
When the historical fault data and the historical normal data are screened and divided in a standard sampling mode, the detailed state of the power grid corresponding to the time of the historical fault data and the time of the historical normal data is obtained, and when the fault data time sequence and the normal data time sequence are obtained through division, the detailed state of the power grid corresponding to the time in the fault data time sequence and the normal data time sequence is associated.
The detailed power grid state corresponding to the time of the historical fault data and the historical normal data identifies the specific state of the equipment of the booster station at the corresponding moment, for example, in the fault time, the fault units are different, so that the specific equipment needs to be marked separately, and each historical fault data and each historical normal data can correspond to the specific equipment node state; therefore, in the subsequent training, the method has more detailed state labeling, and the finally obtained model has more accurate booster station fault recognition effect.
S2, dividing a plurality of monitoring data time sequences into a monitoring data time sequence sample set and a monitoring data time sequence test set;
selecting a fault data time sequence sample set and a fault data time sequence test set from the obtained fault data time sequences according to a set selection proportion; selecting a normal data time sequence sample set and a normal data time sequence test set from the obtained normal data time sequences according to a set selection proportion;
the ratio of the historical data time series sample set to the historical data time series test set is 4:1 when dividing, and the historical data time series test set is set according to the existing historical monitoring data quantity when dividing the ratio actually so as to obtain a more accurate monitoring model.
S3, constructing a digital twin body data model of the transformer substation through basic data in the booster station, substituting a monitoring data time sequence sample set into the digital twin body data model, and training to obtain a primary training digital twin body data model;
and substituting the fault data time series sample set and the normal data time series sample set into training respectively when training the digital twin volume data model.
The basic data in the booster station includes: the physical nodes of all the devices in the booster station are provided with attribute data, topological connection relation, device operation data and environment data, so that the constructed digital twin body data model can be more attached to an actual booster station, and a more accurate early warning effect is achieved in the follow-up intelligent active early warning.
S4, testing the initial training digital twin body data model by adopting a monitoring data time sequence test set, and performing parameter debugging in the test until the test passes, so as to obtain a mature digital twin body data model;
s5, after the mature digital twin body data model is obtained, constructing an AI digital control platform, and constructing a mature digital twin body database model into an operation framework of the AI digital control platform to obtain the AI digital control platform capable of operating the mature digital twin body database model;
and S6, after the AI digital control platform is obtained, accessing the online monitoring data of the booster station, inputting the online monitoring data into the AI digital control platform, running the mature digital twin database model in real time to calculate, obtaining simulation result data, and if the simulation result data is matched with a preset early warning data template, actively starting a corresponding early warning scheme.
After the AI digital control platform is put into use, the evaluation data of simulation result data are obtained by a user at regular time, and the mature digital twin body data model is subjected to parameter optimization according to the evaluation data, namely, after the AI digital control platform is put into use, the model training is carried out by using the operation data of the booster station, so that the training set of the digital twin body data model is larger, more accurate model parameters are obtained, and further, a better early warning effect is obtained when the digital twin body data model is applied.
After the optimized parameters are calculated by the mature digital twin data model, the artificial approval process is carried out, the artificial approval result is obtained, the parameters used by the mature digital twin data model are determined according to the artificial approval result, and after the parameters of the mature digital twin data model obtain more optimized values, a manual approval control link is added, so that the parameters of the digital twin data model are more reliable.
Corresponding to the method, the application also provides a booster station active early warning system based on the AI time sequence twinning technology, which comprises the following steps:
the historical monitoring data processing module is used for acquiring historical monitoring data of the booster station and processing the historical monitoring data to obtain a plurality of monitoring data time sequences with set sample time lengths;
the training set dividing module divides the plurality of monitoring data time sequences into a monitoring data time sequence sample set and a monitoring data time sequence test set;
the digital twin body data model training module is used for constructing a digital twin body data model of the transformer substation through basic data in a booster station, substituting a monitoring data time sequence sample set into the digital twin body data model, and training to obtain a primary training digital twin body data model;
the digital twin body data model test module adopts a monitoring data time sequence test set to test the initial training digital twin body data model, and performs parameter debugging in the test until the test passes, so as to obtain a mature digital twin body data model;
the AI digital control platform module is used for constructing an AI digital control platform after the mature digital twin body data model is obtained, and constructing the mature digital twin body database model into an operation framework of the AI digital control platform to obtain the AI digital control platform capable of operating the mature digital twin body database model;
and the active early warning module is used for accessing the online monitoring data of the booster station after the AI digital control platform is obtained, inputting the online monitoring data into the AI digital control platform to run the mature digital twin database model in real time for calculation to obtain simulation result data, and actively starting a corresponding early warning scheme if the simulation result data is matched with a preset early warning data template.
The booster station active early warning method and system based on the AI time sequence twin technique have the advantages that: when the historical monitoring data are acquired, data are acquired in a distinguishing mode according to the state of the booster station, the historical fault data of the booster station in a fault state and the historical normal data of the booster station in a normal operation state are respectively acquired, and meanwhile, the historical monitoring data in the two states are respectively processed to obtain time sequence training sets in the two states. The digital twin body data model is trained by utilizing the time sequence training sets of the two states respectively to obtain the parameters of the digital twin body data model attached to the booster station to be served currently, and the digital twin body data model is built according to the booster station to be served, so that more accurate simulation result data can be provided during actual service, and further more accurate early warning effect is provided during follow-up intelligent active early warning.
It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor referred to in the embodiments provided in the present application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, or the like, but is not limited thereto.
The above description is only a preferred embodiment of the present application, and the protection scope of the present application is not limited to the above examples, and all technical solutions belonging to the concept of the present application belong to the protection scope of the present application. It should be noted that modifications and adaptations to the present application may occur to one skilled in the art without departing from the principles of the present application and are intended to be within the scope of the present application.

Claims (9)

1. The booster station active early warning method based on the AI time sequence twinning technology is characterized by comprising the following steps: the method comprises the following steps:
acquiring historical monitoring data of the booster station, and processing the historical monitoring data to obtain a plurality of monitoring data time sequences with set sample time lengths;
dividing a plurality of monitoring data time sequences into a monitoring data time sequence sample set and a monitoring data time sequence test set;
constructing a digital twin body data model of the transformer substation through basic data in the booster station, substituting a monitoring data time sequence sample set into the digital twin body data model, and training to obtain a primary training digital twin body data model;
testing the initial training digital twin body data model by adopting a monitoring data time sequence test set, and performing parameter debugging in the test until the test passes, so as to obtain a mature digital twin body data model;
after the mature digital twin body data model is obtained, an AI digital control platform is constructed, and the mature digital twin body database model is constructed into an operation framework of the AI digital control platform to obtain the AI digital control platform capable of operating the mature digital twin body database model;
after the AI digital control platform is obtained, the online monitoring data of the booster station is accessed, the online monitoring data is input into the AI digital control platform to run the mature digital twin database model in real time for calculation, the simulation result data is obtained, and if the simulation result data is matched with a preset early warning data template, the corresponding early warning scheme is actively started.
2. The booster station active early warning method based on the AI time sequence twinning technology of claim 1, characterized by comprising the following steps: after the historical monitoring data of the booster station is obtained, the historical monitoring data is processed, and the method comprises the following steps of: and carrying out structuring treatment on the obtained historical monitoring data, and extracting multidimensional time series data of each monitoring item.
3. The booster station active early warning method based on the AI time sequence twinning technology as defined in claim 2, wherein the method is characterized by comprising the following steps: when the historical monitoring data of the booster station are obtained, the historical fault data of the booster station in a fault state and the historical normal data of the booster station in a normal running state are respectively obtained; screening and dividing the historical fault data and the historical normal data in a standard sampling manner respectively to obtain a plurality of continuous fault data time sequences and normal data time sequences with set sample time lengths;
selecting a fault data time sequence sample set and a fault data time sequence test set from the obtained fault data time sequences according to a set selection proportion; selecting a normal data time sequence sample set and a normal data time sequence test set from the obtained normal data time sequences according to a set selection proportion;
and substituting the fault data time series sample set and the normal data time series sample set into training respectively when training the digital twin volume data model.
4. The booster station active early warning method based on the AI time sequence twinning technology as defined in claim 3, wherein the method is characterized by comprising the following steps: when the historical fault data and the historical normal data are screened and divided in a standard sampling mode, the detailed state of the power grid corresponding to the time of the historical fault data and the time of the historical normal data is obtained, and when the fault data time sequence and the normal data time sequence are obtained through division, the detailed state of the power grid corresponding to the time in the fault data time sequence and the normal data time sequence is associated.
5. The booster station active early warning method based on the AI time sequence twinning technology of claim 4, wherein the method comprises the following steps: the ratio of the historical data time series sample set to the historical data time series test set at the time of division is 4:1.
6. The booster station active early warning method based on the AI time sequence twinning technology of claim 5, wherein the method comprises the following steps: the basic data in the booster station comprises: attribute data, topological connection relation, equipment operation data and environment data of physical nodes where all equipment in the booster station are located.
7. The booster station active early warning method based on the AI time sequence twinning technology of claim 6, wherein the method comprises the following steps: after the AI digital control platform is put into use, the evaluation data of the simulation result data of the user is obtained at regular time, and the mature digital twin body data model is subjected to parameter optimization according to the evaluation data.
8. The booster station active early warning method based on the AI time sequence twinning technique of claim 7, wherein the method comprises the following steps: after the optimized parameters are calculated by the mature digital twin volume data model, the artificial approval process is carried out, the artificial approval result is obtained, and the parameters used by the mature digital twin volume data model are determined according to the artificial approval result.
9. A booster station active early warning system based on an AI time sequence twinning technology is characterized in that: comprising the following steps:
the historical monitoring data processing module is used for acquiring historical monitoring data of the booster station and processing the historical monitoring data to obtain a plurality of monitoring data time sequences with set sample time lengths;
the training set dividing module divides the plurality of monitoring data time sequences into a monitoring data time sequence sample set and a monitoring data time sequence test set;
the digital twin body data model training module is used for constructing a digital twin body data model of the transformer substation through basic data in a booster station, substituting a monitoring data time sequence sample set into the digital twin body data model, and training to obtain a primary training digital twin body data model;
the digital twin body data model test module adopts a monitoring data time sequence test set to test the initial training digital twin body data model, and performs parameter debugging in the test until the test passes, so as to obtain a mature digital twin body data model;
the AI digital control platform module is used for constructing an AI digital control platform after the mature digital twin body data model is obtained, and constructing the mature digital twin body database model into an operation framework of the AI digital control platform to obtain the AI digital control platform capable of operating the mature digital twin body database model;
and the active early warning module is used for accessing the online monitoring data of the booster station after the AI digital control platform is obtained, inputting the online monitoring data into the AI digital control platform to run the mature digital twin database model in real time for calculation to obtain simulation result data, and actively starting a corresponding early warning scheme if the simulation result data is matched with a preset early warning data template.
CN202310760831.8A 2023-06-27 2023-06-27 Boosting station active early warning method and system based on AI time sequence twinning technology Pending CN117040106A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310760831.8A CN117040106A (en) 2023-06-27 2023-06-27 Boosting station active early warning method and system based on AI time sequence twinning technology

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310760831.8A CN117040106A (en) 2023-06-27 2023-06-27 Boosting station active early warning method and system based on AI time sequence twinning technology

Publications (1)

Publication Number Publication Date
CN117040106A true CN117040106A (en) 2023-11-10

Family

ID=88640097

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310760831.8A Pending CN117040106A (en) 2023-06-27 2023-06-27 Boosting station active early warning method and system based on AI time sequence twinning technology

Country Status (1)

Country Link
CN (1) CN117040106A (en)

Similar Documents

Publication Publication Date Title
CN112597691B (en) LightGBM algorithm-based fault early warning method for wind turbine generator variable pitch motor temperature sensor
CN110570122A (en) Offshore wind power plant reliability assessment method considering wind speed seasonal characteristics and current collection system element faults
CN111415059B (en) Practical model machine construction and online application method
CN114742283A (en) Medium-voltage line loop closing current estimation and risk assessment method based on deep learning network
CN117669361A (en) Power distribution area topology identification function verification method and system
CN114186858A (en) Method and system for evaluating importance of energy storage planning node of power distribution network
CN111177278A (en) Grid user short-term load prediction real-time processing tool
CN114336628A (en) Photovoltaic power supply access distribution network safety evaluation method based on random power flow
Li et al. A line-fault cause analysis method for distribution network based on decision-making tree and machine learning
CN115469184A (en) New energy transmission line fault identification method based on convolutional network
CN114997566A (en) Power grid blocking risk assessment method and system considering node connectivity loss
CN113850017A (en) System-level fault analysis system and method based on power flow change map
CN113468745A (en) Power distribution network reliability rapid evaluation method and system based on historical faults
CN115224684A (en) Intelligent power distribution network risk state identification method and system based on immune hazard theory
CN117422299A (en) Risk assessment method for power transmission and transformation system containing multi-state wind power plant
CN117040106A (en) Boosting station active early warning method and system based on AI time sequence twinning technology
CN115809836A (en) Distribution network toughness planning method considering distributed energy storage emergency power supply capacity
CN116776232A (en) Peak-shaving power supply equipment data mining method
CN105373862A (en) Terminal station planning scheme reliability evaluation optimal selection method based on annual load curve
CN110797906B (en) Method and device for determining maximum installed capacity of wind power base and storage medium
Bi et al. Explainable Artificial Intelligence for Power System Security Assessment: A Case Study on Short-Term Voltage Stability
Wang et al. On the Correlation Model between Extreme Disaster Weather and Distribution Transformer Fault Types
CN106777494B (en) Method for calculating sensitivity of reliability influence factors of power system
Zhou et al. Power Grid transient stability prediction method based on improved CNN under big data background
CN117559495B (en) Power distribution network energy storage planning method and device based on service condition of storage battery

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