CN116914921A - Power grid running state determining method, device, equipment and storage medium - Google Patents

Power grid running state determining method, device, equipment and storage medium Download PDF

Info

Publication number
CN116914921A
CN116914921A CN202310721388.3A CN202310721388A CN116914921A CN 116914921 A CN116914921 A CN 116914921A CN 202310721388 A CN202310721388 A CN 202310721388A CN 116914921 A CN116914921 A CN 116914921A
Authority
CN
China
Prior art keywords
power grid
fault
current signal
running state
determining
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
CN202310721388.3A
Other languages
Chinese (zh)
Inventor
孙昕杰
王树彪
申张亮
荆江平
张刘冬
秦涛
张昊亮
赵浩君
吴昊
董丹丹
刘思嘉
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.)
Nanjing Power Supply Co of State Grid Jiangsu Electric Power Co Ltd
Original Assignee
Nanjing Power Supply Co of State Grid Jiangsu Electric Power 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 Nanjing Power Supply Co of State Grid Jiangsu Electric Power Co Ltd filed Critical Nanjing Power Supply Co of State Grid Jiangsu Electric Power Co Ltd
Priority to CN202310721388.3A priority Critical patent/CN116914921A/en
Publication of CN116914921A publication Critical patent/CN116914921A/en
Pending legal-status Critical Current

Links

Classifications

    • 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
    • H02J13/00002—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 characterised by monitoring
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
    • G01R19/0092—Arrangements for measuring currents or voltages or for indicating presence or sign thereof measuring current only
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/08—Locating faults in cables, transmission lines, or networks
    • G01R31/081—Locating faults in cables, transmission lines, or networks according to type of conductors
    • G01R31/086—Locating faults in cables, transmission lines, or networks according to type of conductors in power transmission or distribution networks, i.e. with interconnected conductors
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/08—Locating faults in cables, transmission lines, or networks
    • G01R31/088—Aspects of digital computing
    • 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
    • G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
    • 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
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G06F18/253—Fusion techniques of extracted features
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/048—Activation functions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • 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
    • H02J13/00001—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 characterised by the display of information or by user interaction, e.g. supervisory control and data acquisition systems [SCADA] or graphical user interfaces [GUI]
    • 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
    • H02J13/00006—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 characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment
    • H02J13/00016—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 characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment using a wired telecommunication network or a data transmission bus
    • 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
    • H02J13/00006—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 characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment
    • H02J13/00022—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 characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment using wireless data transmission
    • 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
    • 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)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Mathematical Physics (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Probability & Statistics with Applications (AREA)
  • Human Computer Interaction (AREA)
  • Testing And Monitoring For Control Systems (AREA)

Abstract

The embodiment of the invention discloses a method, a device, equipment and a storage medium for determining the running state of a power grid, which comprise the following steps: acquiring current signal data and fault term data; inputting current signal data and fault term data into a pre-trained power grid running state identification model; the power grid running state identification model comprises a cross correlation module for performing characteristic cross correlation processing on current signal data and fault term data; and determining the running state of the power grid according to the output result of the power grid running state identification model. The method has the advantages that fusion of current data and semantic data in the power grid is realized, the determination of the running state of the power grid is realized from the comprehensive direction of multiple dimensions, the utilization rate of the collected data in the running state determination of the power grid in the running process of the power grid is improved, the accuracy of the running state determination of the power grid is improved, and the fault diagnosis performance of the power grid is improved.

Description

Power grid running state determining method, device, equipment and storage medium
Technical Field
The present invention relates to the field of power grid operation and maintenance technologies, and in particular, to a method, an apparatus, a device, and a storage medium for determining a power grid operation state.
Background
Along with the popularization of the power grid informatization system, the data for power grid fault diagnosis are more abundant and diversified, and diagnosis of the power grid faults by using diversified information is expected to improve diagnosis accuracy.
However, the grid diversification data has granularity, semantic differences and heterogeneous characteristics, such as three-phase current data for representing current information and text information for representing fault anomalies can exist in the system at the same time when the grid informatization system monitors the grid operation information.
When the running state of the power grid is determined or monitored, fault judgment is often carried out only based on current data acquired in the system, full utilization of the diversified data is difficult to achieve, fusion of the multiple data is difficult, fault diagnosis errors of the power grid are often caused by forced fusion, and operation and maintenance difficulty of the power grid is improved.
Disclosure of Invention
The invention provides a method, a device, equipment and a storage medium for determining the running state of a power grid, which are characterized in that a power grid running state identification model with a cross-correlation module is used for fusing current signal data and fault term data acquired during monitoring the running state of the power grid, and then the power grid fault is diagnosed according to the fused characteristic data so as to determine the running state of the power grid, the running state of the power grid is comprehensively determined by utilizing multivariate information, the accuracy of determining the running state of the power grid is improved, and the fault diagnosis performance of the power grid is improved.
In a first aspect, an embodiment of the present invention provides a method for determining an operating state of a power grid, including:
acquiring current signal data and fault term data;
inputting current signal data and fault term data into a pre-trained power grid running state identification model; the power grid running state identification model comprises a cross correlation module for performing characteristic cross correlation processing on current signal data and fault term data;
and determining the running state of the power grid according to the output result of the power grid running state identification model.
In a second aspect, an embodiment of the present invention further provides a power grid operation state determining device, including:
the data acquisition module is used for acquiring current signal data and fault term data;
the model processing module is used for inputting the current signal data and the fault term data into a pre-trained power grid running state identification model; the power grid running state identification model comprises a cross correlation module for performing characteristic cross correlation processing on current signal data and fault term data;
and the running state determining module is used for determining the running state of the power grid according to the output result of the power grid running state identification model.
In a third aspect, an embodiment of the present invention further provides a power grid operation state determining device, including:
At least one processor; and
a memory communicatively coupled to the at least one processor; wherein, the liquid crystal display device comprises a liquid crystal display device,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the grid operation state determination method provided by the embodiment of the present invention.
In a fourth aspect, embodiments of the present invention also provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform the grid operation state determination method provided by the embodiments of the present invention.
The embodiment of the invention provides a method, a device, equipment and a storage medium for determining the running state of a power grid, which are used for acquiring current signal data and fault term data; inputting current signal data and fault term data into a pre-trained power grid running state identification model; the power grid running state identification model comprises a cross correlation module for performing characteristic cross correlation processing on current signal data and fault term data; and determining the running state of the power grid according to the output result of the power grid running state identification model. By adopting the technical scheme, when the power grid faults are diagnosed through the data collected in the power grid monitoring system, and further the monitoring of the power grid running state is realized, the current signal data of the electric signal type and the fault term data of the text type collected in the system are input into the pre-trained power grid running state identification model to be processed, the cross-correlation processing is carried out on the characteristics extracted from the current signal data and the fault term data through the cross-correlation module in the power grid running state identification model, the output result of the power grid running state identification model is obtained according to the characteristic information after the cross-correlation processing, and then the power grid running state is determined according to the output result of the model. The method has the advantages that various power grid operation information acquired in the power grid informatization system can be fully utilized to judge the power grid operation state, the problem that diversified data are difficult to comprehensively represent the same power grid fault due to semantic granularity difference, heterogeneous characteristics and the like is solved, fusion of current data and semantic data in the power grid is realized, the determination of the power grid operation state is realized from the comprehensive direction of multiple dimensions, the utilization rate of the acquired data in the power grid operation state determination in the power grid operation process is improved, the accuracy of the power grid operation state determination is improved, and the power grid fault diagnosis performance is improved.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for determining an operation state of a power grid according to a first embodiment of the present invention;
fig. 2 is a flowchart of a method for determining an operation state of a power grid according to a second embodiment of the present invention;
fig. 3 is a diagram illustrating a structure of a current feature extraction module according to a second embodiment of the present invention;
fig. 4 is a flowchart illustrating a process of inputting current signal data to a current feature extraction module for feature extraction and determining current signal features according to a second embodiment of the present invention;
fig. 5 is a flowchart illustrating a process of inputting current signal features and text semantic features to a cross correlation module for cross correlation processing and determining fusion features according to a second embodiment of the present invention;
Fig. 6 is a schematic structural diagram of a power grid operation state determining device according to a third embodiment of the present invention;
fig. 7 is a schematic structural diagram of a power grid operation state determining device according to a fourth embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
Fig. 1 is a flowchart of a method for determining an operating state of a power grid according to a first embodiment of the present invention, where the method may be performed by a power grid operating state determining device, which may be implemented by software and/or hardware, and the power grid operating state determining device may be configured in a power grid operating state determining device, where the method is applicable to a case of determining an operating state of a power grid based on multiple types of power grid monitoring data collected in a power grid informatization system. Alternatively, the power grid operation device may be an electronic device, where the electronic device may be a notebook, a desktop computer, or an intelligent tablet, and the embodiment of the present invention is not limited thereto.
As shown in fig. 1, the method for determining the running state of the power grid provided by the embodiment of the invention specifically includes the following steps:
s101, acquiring current signal data and fault term data.
In this embodiment, the current signal data may be specifically understood as three-phase current data in each node or each line of the power grid, which is collected by the grid informatization system during operation of the power grid. The fault term data can be specifically understood as text information which is acquired by a power grid informatization system in the running process of the power grid or is input into the power grid informatization system after the acquired information is analyzed by a technician and used for indicating that a fault abnormality exists in a certain node or a certain line in the power grid.
Specifically, when the running state of the power grid needs to be determined, the power grid area and the time for determining the running state can be determined according to actual requirements, three-phase current data and text information indicating fault abnormality in the power grid area in the time are obtained from a power grid informatization system, and the three-phase current data and the text information are respectively used as current signal data and fault term data.
S102, inputting current signal data and fault term data into a pre-trained power grid running state identification model.
The power grid running state identification model comprises a cross correlation module used for carrying out characteristic cross correlation processing on current signal data and fault term data.
In this embodiment, the power grid operation state recognition model may be specifically understood as performing feature extraction, cross-correlation fusion, classification and other processes on the current signal data and the fault term data input therein, so as to generate a neural network model including the output result of the power grid fault information. The cross-correlation module can be specifically understood as a plurality of neural network layers used for determining the cross-correlation degree between the characteristics of the current signal data and the fault function term data which belong to different types in the power grid running state identification model, and further completing the fusion of the two characteristics according to the cross-correlation degree.
Specifically, current signal data and fault term data are input into a pre-trained power grid running state model, feature extraction is carried out on the current signal data and the fault term data respectively to obtain corresponding current signal features and text semantic features, the current signal features and the text semantic features are further input into a cross correlation module together, the cross correlation degree between the current signal features and the text semantic features is determined, feature fusion is carried out on the current signal features and the text semantic features according to the cross correlation degree, and the fused features are classified by utilizing a full connection layer and a pooling layer in a power grid running state identification model to obtain an output result of the power grid running state model.
S103, determining the running state of the power grid according to the output result of the power grid running state identification model.
In this embodiment, the operating state of the power grid may be specifically understood as information indicating whether or not there is an operation failure in each node and line in the power grid.
Specifically, since the output result of the power grid running state model is determined according to the current signal data and the fault term data, and the power grid area and the fault label which possibly has faults in the time required to be determined for the running state are determined according to the actual demand, each fault label can be analyzed to determine the node or the line which possibly has faults in the power grid and the corresponding faults in the fault node or the fault line, and the information is integrated to be the power grid running state.
According to the technical scheme, current signal data and fault term data are obtained; inputting current signal data and fault term data into a pre-trained power grid running state identification model; the power grid running state identification model comprises a cross correlation module for performing characteristic cross correlation processing on current signal data and fault term data; and determining the running state of the power grid according to the output result of the power grid running state identification model. By adopting the technical scheme, when the power grid faults are diagnosed through the data collected in the power grid monitoring system, and further the monitoring of the power grid running state is realized, the current signal data of the electric signal type and the fault term data of the text type collected in the system are input into the pre-trained power grid running state identification model to be processed, the cross-correlation processing is carried out on the characteristics extracted from the current signal data and the fault term data through the cross-correlation module in the power grid running state identification model, the output result of the power grid running state identification model is obtained according to the characteristic information after the cross-correlation processing, and then the power grid running state is determined according to the output result of the model. The method has the advantages that various power grid operation information acquired in the power grid informatization system can be fully utilized to judge the power grid operation state, the problem that diversified data are difficult to comprehensively represent the same power grid fault due to semantic granularity difference, heterogeneous characteristics and the like is solved, fusion of current data and semantic data in the power grid is realized, the determination of the power grid operation state is realized from the comprehensive direction of multiple dimensions, the utilization rate of the acquired data in the power grid operation state determination in the power grid operation process is improved, the accuracy of the power grid operation state determination is improved, and the power grid fault diagnosis performance is improved.
Example two
Fig. 2 is a flowchart of a power grid operation state determining method provided by a second embodiment of the present invention, where the technical solution of the embodiment of the present invention is further optimized based on the above-mentioned alternative technical solutions, and the current signal data is extracted by a current feature extraction module including a shared learning layer, so that a multi-scale combined architecture for extracting the current signal data features may be optimized, and further, the extracted current features are the best multi-scale fusion current features. When semantic feature extraction is carried out on fault term data, the cutoff length of the fault term data is the same as the current feature length, so that granularity difference between the extracted current signal features and text semantic features is reduced, and correlation calculation and cross correlation fusion are conveniently carried out on the fault term data by a cross correlation module. And the output result is obtained by classifying the cross-correlation fused features, and possible faults in the power grid are determined according to the number of fault labels in the output result and the probability of the fault labels corresponding to the fault labels, so that the running state of the power grid is determined, the utilization rate of the acquired data in the running process of the power grid in the running state determination of the power grid is improved, and the accuracy of the running state determination of the power grid is improved.
In this embodiment, the power grid operation state identification model further includes: the device comprises a semantic feature extraction module, a current feature extraction module and a feature classification module. The semantic feature extraction module is specifically understood as a set of neural network layers for extracting semantic features of the fault term data input therein. The current feature extraction module is specifically understood as a neural network layer set constructed according to a multi-scale shared learning mechanism and used for performing multi-scale feature extraction and fusion on current signal data input into the current feature extraction module. The feature classification module may be understood as a set of neural network layers for data adoption and classification of features entered therein to determine the probability that various types of fault labels may exist based on the input features.
As shown in fig. 2, a method for determining an operation state of a power grid according to a second embodiment of the present invention specifically includes the following steps:
s201, acquiring current signal data and fault term data.
S202, inputting the current signal data to a current feature extraction module for feature extraction, and determining current signal features.
The current characteristic extraction module comprises a multi-scale coarse grain layer, a first convolution layer, a first shared learning layer, a second convolution layer, a second shared learning layer and a first full-connection layer.
The convolution kernels arranged in the first convolution layer and the second convolution layer are matched with each coarse grain scale factor.
In this embodiment, the multi-scale coarse-grained layer may be specifically understood as a neural network layer for sampling current signal data according to different scales to obtain current signal data with different scales and different granularities, so as to extract current characteristics at different scales. Further, the multi-scale coarse-grained layer can complete sampling of the current signal data according to different coarse-grained scale factors, and the coarse-grained scale factors can be specifically understood as time scale factors set according to actual requirements. The first convolution layer, the second convolution layer and the third convolution layer are specifically understood as a neural network layer formed by a plurality of different convolution kernels and used for carrying out local area feature extraction on information input into the neural network layer. The first shared learning layer and the second shared learning layer can be understood as a weight matrix formed by weights for optimizing the multi-scale combination architecture of the current feature extraction module.
Fig. 3 is a diagram illustrating a structure of a current feature extraction module according to a second embodiment of the present invention, and fig. 4 is a diagram illustrating a process of inputting current signal data to the current feature extraction module to perform feature extraction and determining current signal features, as shown in fig. 4, and specifically includes the following steps:
s2021, inputting current signal data into the multi-scale coarse-grained layer, and generating a current information sequence corresponding to each coarse-scale factor according to a preset number of coarse-scale factors.
Specifically, the multi-scale coarse-grained layer may include a plurality of coarse-grained scale factors set according to actual situations, and when current signal data is input, each coarse-grained scale factor may be utilized to sample and extract characteristics of the current signal data, and then the coarse-grained scale factors are used to average the current signal data, so as to obtain a current information sequence corresponding to the coarse-grained scale factors, and thus a current information sequence corresponding to each coarse-grained scale factor may be generated.
Illustratively, assume that the coarse-grained scale factor is expressed asτ, the current signal data is represented as x (t) = { x 1 (t),x 2 (t),...,x n (t) }, n is the number of current data segments sampled by the grid informatization system, x l (t) is the first data segment in the current signal data x (t), and t is the time of the time domain signal. In the current signal data x (t), each time of jump tau data segments, taking tau as an interval scale to average to form a current information sequence with the current signal data time scale tau, which can be specifically expressed by the following formula:
wherein, the liquid crystal display device comprises a liquid crystal display device,the current information sequence with the h-th coarse-grain scale factor tau obtained through coarse-grain calculation is optional, in the embodiment of the invention, coarse-grain scale factors tau can be selected from 1, 3 and 5, and the coarse-grain scale factors can be adaptively set according to actual requirements, which is not limited in the embodiment of the invention.
S2022, inputting each current information sequence into a first convolution layer to perform convolution operation, and determining a first output characteristic corresponding to each current information sequence.
In the above example, since coarse-grained scale factors corresponding to different current information sequences are different, in order to adapt to convolution processing of different scale data, a convolution kernel used for convolution processing of each current information sequence in the first convolution layer should be adapted to the coarse-grained scale factor corresponding to the current information sequence. In the embodiment of the invention, for a current information sequence with a coarse-grained scale factor of tau=1, a convolution kernel with a step length of 1 and a scale of 7 is adopted for adaptation; aiming at a current information sequence with a coarse-grain scale factor of tau=3, adopting a convolution kernel with a step length of 1 and a scale of 6 for adaptation; for a current information sequence with a coarse-grained scale factor of τ=5, a convolution kernel with a step size of 1 and a scale of 4 is used for adaptation. And finally outputting first output characteristics corresponding to each current information sequence, wherein the lengths of the first output characteristics are N-4, and N is the length of the current information sequence.
S2023, inputting each first output feature to the first shared learning layer, so that the first output feature matrix formed by each first output feature is multiplied by the first shared matrix in the first shared learning layer, and the first shared output feature corresponding to each first output feature is determined.
Specifically, since the first shared learning layer includes a first shared matrix formed by a plurality of weights, in order to optimize the multi-scale combination architecture, each first output feature may be formed into a first output feature matrix corresponding to the first shared matrix, the first shared matrix is multiplied by the first output feature matrix, each first output feature in the first output feature matrix is weighted by the weight in the first shared matrix, so as to obtain the first shared output feature matrix, and the first shared output feature matrix may include the first shared output feature corresponding to each first output feature scale.
Following the above example, assume that the first output characteristic with coarse-grained scale factor τ=1 is f 1 1 The first output characteristic of a coarse-grained scale factor τ=3 is f 2 1 The first output characteristic of coarse-grained scale factor τ=5 is f 3 1 The first shared matrix is expressed asThe first output characteristic matrix corresponding to the first shared matrix is expressed as +. >The corresponding first shared output feature matrix may be represented by:
wherein, the liquid crystal display device comprises a liquid crystal display device,is equal to the first output characteristic f 1 1 First shared output feature with same coarse-grained scale factor,/->For the first output characteristic->First shared output feature with same coarse-grained scale factor,/->Is equal to the first output characteristic f 3 1 The coarse-grained scale factor is the same as the first shared output feature.
S2024, inputting each first shared output characteristic into a second convolution layer to perform convolution operation, and determining a second output characteristic corresponding to each first shared output characteristic.
Specifically, similar to the first convolution layer, when the convolution processing is performed on the first shared output features of different coarse-grained scale factors in the second convolution layer, the adaptation processing is performed by adopting the same convolution kernel as that in the first convolution layer, and the second output features corresponding to the coarse-grained scale factors of each first shared output feature are obtained after the convolution processing.
S2025, inputting each second output feature to the second shared learning layer, so that a second output feature matrix formed by each second output feature is multiplied by a second shared matrix in the second shared learning layer, and a second shared output feature corresponding to each second output feature is determined.
Specifically, similar to the first shared learning layer, the second shared learning layer also includes a second shared matrix formed by a plurality of weights, in order to optimize a multi-scale combination architecture for fusing the multi-scale current signal features, each second output feature is formed into a second output feature matrix corresponding to the second shared matrix, the second shared matrix is multiplied by the second output feature matrix, each second output feature in the second output feature matrix is weighted by the weight in the second shared matrix, so as to obtain a second shared output feature matrix, and the second shared output feature matrix may include second shared output features corresponding to each second output feature scale.
Following the above example, assume that the second output characteristic with a coarse-grained scale factor of τ=1 after convolution processing with the second convolution layer is f 1 2 The second output characteristic of coarse-grained scale factor τ=3 isThe second output characteristic of coarse-grained scale factor τ=5 is f 3 2 The second shared matrix is denoted +.>Each of the second output characteristic formations corresponding to the second shared matrix may be expressed as +.>The corresponding second shared output feature matrix may be represented by:
Wherein, the liquid crystal display device comprises a liquid crystal display device,is equal to the second output characteristic f 1 2 A second shared output feature with the same coarse-grained scale factor,/->For the second output characteristic->A second shared output feature with the same coarse-grained scale factor,/->Is equal to the second output characteristic f 3 2 Second sharing of coarse-grained scale factorsAnd outputting the characteristics.
S2026, inputting the second shared output features into a third convolution layer to perform convolution operation, and determining third output features corresponding to the second shared output features.
Specifically, similar to the first convolution layer, when the second shared output features with different coarse-grained scale factors are convolved in the third convolution layer, the same convolution kernel as that in the first convolution layer is adopted to perform adaptation processing, and after the convolution processing, the third output features corresponding to the coarse-grained scale factors of the second shared output features are obtained.
S2027, inputting the third output characteristics into the full-connection layer for characteristic fusion, and determining the current signal characteristics.
Specifically, third output characteristics corresponding to different coarse-grained scale factors are input to the full-connection layer, fusion is carried out in a mode such as a concat function, and characteristics obtained after fusion are determined to be current signal characteristics.
S203, fault term data are input to a semantic feature extraction module to perform feature extraction, and text semantic features are determined.
Specifically, the fault term data is input into a semantic feature extraction module for extracting semantic features of characters, and the feature extracted by the fault term data is determined as text semantic features because the current signal features and the text semantic features can be fused with smaller barriers, and the cutoff length of the semantic feature extraction can be set according to the length of the current information sequence when the semantic extraction task is executed. The text semantic features and the current signal features are consistent in data length, the granularity difference of the text semantic features and the current signal features is reduced, and the subsequent cross correlation module can conveniently determine and fuse the cross correlation of the two features.
Optionally, the semantic feature extraction module in the embodiment of the present invention may be a long-short-term memory network, or may be a set of other neural network layers with semantic feature extraction functions, which is not limited in the embodiment of the present invention.
It is understood that S202 and S203 may be performed simultaneously, or may be performed in any order, and in the embodiment of the present invention, S202 and S203 are performed simultaneously as an example.
S204, inputting the current signal characteristics and the text semantic characteristics into a cross correlation module for cross correlation processing, and determining fusion characteristics.
Specifically, the current signal features and the text semantic features are input into the cross-correlation module, cross-correlation between the current signal features and the text semantic features is determined by using a cross-correlation function contained in the cross-correlation module, namely, the cross-correlation degree between two different features is determined, and then feature fusion is carried out on the current signal features and the text semantic features according to the cross-correlation degree, and the generated features are determined to be fusion features.
Further, fig. 5 is a flowchart illustrating a process of inputting current signal features and text semantic features to a cross-correlation module for cross-correlation processing and determining fusion features, as shown in fig. 5, and specifically includes the following steps:
s2041, inputting the current signal features and the text semantic features into a cross correlation module, and determining the maximum average difference distance between the current signal features and the text semantic features.
Specifically, the current signal features and the text semantic features are input to the cross correlation module, and as the current signal features and the text semantic features are two different but related variables, the distances of the two distributions can be characterized by the maximum average difference distance (Maximum Mean Discrepancy, MMD).
S2042, fusing the current signal characteristics and the text semantic characteristics by minimizing the maximum average difference distance, and generating fused characteristics.
Specifically, in order to enable the current signal features and text semantic features with different properties to be better fused, and further possible faults in the operation of the power grid can be determined according to the features containing comprehensive information after fusion, the current signal features and the text semantic features can be fused when the correlation degree of the current signal features and the text semantic features is maximum according to the idea of minimizing the maximum average difference distance, and fusion features are obtained.
S205, inputting the fusion features into a feature classification module to perform fault division, and determining the determined fault label probability as an output result of a power grid running state identification model.
Alternatively, the feature classification module may include a full connection layer and a global pooling layer.
Specifically, since the fusion feature includes the current signal feature and the text semantic feature in the same node or the same line stage in the power grid, the current signal feature and the text semantic feature can be combined to determine the power grid fault condition existing in the node or the line stage, when the power grid running state identification model is trained, a plurality of fault labels with different fault types are set for the power grid running state identification model, after the fusion feature is input into the feature classification module, the fusion feature can be divided according to the fault types through the full connection layer and the global pooling layer in the feature classification module, the probability that the fusion feature possibly has different fault types is determined, then the fault label probability corresponding to each fault label is output, and each fault label probability is used as the output result of the feature classification module, namely the power grid running state identification model.
Further, before the power grid operation state identification model is put into use, that is, before the current signal data and the fault term data are obtained in the embodiment of the present invention, training of the power grid operation state identification model needs to be completed, and the specific training may include the following steps:
1) And acquiring a historical current signal data set, a historical fault term data set and historical grid faults corresponding to each historical current signal data and the historical fault term data in preset historical time.
In this embodiment, the preset history time may be specifically understood as a period of time before the current time determined according to the actual requirement. A historical current signal data set is understood to mean in particular a set of current signal data stored in the grid information system for a preset historical time. The historical fault term data set can be specifically understood as a set of fault term data which is stored in the power grid informatization system within a preset historical time and is the same as the data acquisition position in the historical current signal data set. It can be understood that, for the same node or line at the same historical moment, there is a correspondence between the historical current signal data and the historical fault term data, and for a set of historical current signal data and historical fault term data, the same fault type contained therein is determined as a historical grid fault corresponding to the set of data.
2) And taking the historical grid faults as tags of the historical current signal data and the historical fault term data, and constructing a training sample set.
Specifically, for a set of historical current signal data and historical fault term data, a historical grid fault corresponding to the set of data is determined as a fault tag corresponding to the set of data, and the set of historical current signal data, the historical fault term data and the fault tag are determined as a training sample. After the combination of each group of historical current signal data and historical fault term data and the fault label determination are completed, the set formed by each training sample is determined to be a training sample set.
3) And inputting the training sample set into an initial power grid running state identification model, constructing a loss function based on the minimum maximum average difference distance, and training the initial power grid running state identification model until a preset stopping condition is met to obtain the power grid running state identification model.
In this embodiment, the initial power grid operation state identification model may be specifically understood as a power grid operation state identification model that is not subjected to weight adjustment, and its architecture is completely consistent with the power grid operation state identification model. The preset stopping condition is specifically understood to be a condition set according to actual situations for determining the convergence state of the power grid operation state identification model, that is, for determining when to stop training for the power grid operation state identification model. Alternatively, the preset stopping condition may be that the training iteration reaches a preset number of times, which is not limited in the embodiment of the present invention.
In the embodiment of the invention, because the cross-correlation module in the power grid running state identification model is a neural network layer set based on the maximum average difference distance, the loss function can be constructed based on the minimum maximum average difference distance in the training process.
Exemplary, assume that the training sample set is input into the initial grid operating state recognition model and is extracted by semantic featuresThe text semantic features obtained after the processing of the fetching module can be expressed as X t The current signal characteristics obtained after the processing of the current characteristic extraction module are expressed as X s Y is a failure tag, then the constructed loss function can be expressed as:
wherein, the liquid crystal display device comprises a liquid crystal display device,for training a current signal characteristic corresponding to the sample set, y i Is->Corresponding fault labels, J () is a cross entropy loss function, λ is a regular term coefficient, D () is a distance measure between a source domain and a target domain, and C () is a classifier.
Because the feature classification module adopts the full connection layer and the global pooling layer to classify the fusion features output by the cross correlation module, the one-dimensional vector obtained by adopting the global pooling layer can be expressed as:
R＝f'η
wherein f' is the weight of the full connection layer, and->Current signal features and text semantic features optimized for minimizing maximum average differences.
The Relu activation function can be expressed as:
in summary, in the embodiment of the present invention, the weight training of each module in the initial power grid operation state identification model may be completed by constructing the loss function based on the minimum maximum average difference distance until the training for the model is considered to be completed when the preset stop condition is satisfied, so as to obtain the power grid operation state identification model that may be used for determining the actual power grid operation state.
S206, determining the number of fault labels contained in the output result and the probability of the fault labels corresponding to the fault labels.
Specifically, after the activation function in the power grid running state identification model is processed, only fault labels with probability larger than zero and fault label probability are output, so that when the number of the fault labels is small, the power grid faults determined according to the current signal data and the fault term data can be considered to be unified, and at the moment, the power grid running state can be determined directly according to the output fault labels. Otherwise, the determined power grid faults in the output result are not uniform, and the probability of each fault label to be output is required to be analyzed so as to determine the running state of the power grid.
S207, if the number of the fault labels is smaller than a preset label number threshold, determining the running state of the power grid according to the corresponding faults of each fault label.
In this embodiment, the preset tag number threshold may be specifically understood as a tag number preset according to an actual situation, which is used to determine that the power grid fault may be directly determined.
Specifically, when the number of fault labels is smaller than a preset label number threshold, the power grid faults determined according to the output result can be considered to be unified, at the moment, the faults corresponding to the fault labels can be determined to be faults of the current signal data and the fault term data in the corresponding region of the power grid, and the faults can be used as the power grid running state of the region.
Further, if no fault label is output, the corresponding region of the artificial current signal data and the fault term data in the power grid has no fault, and at the moment, the power grid running state can be determined to be normal running.
S208, if the number of the fault labels is larger than a preset label number threshold, determining the fault label corresponding to the fault label probability larger than the preset probability threshold as a pending fault label, sorting the pending fault labels according to the fault label probabilities, and determining the running state of the power grid according to the sorting result.
Specifically, when the number of the fault labels is greater than a preset label number threshold, it can be considered that the faults of the power grid determined according to the output result are not uniform, and as the fault label probability of each fault label output by the output result, the part with lower probability can be considered as misjudgment, so that the fault label with the fault label probability greater than the preset probability threshold can be considered as the fault label corresponding to the possible power grid fault at the moment, the fault label obtained by screening is determined to be the pending fault label, and then the pending fault labels obtained by screening are ranked from large to small, wherein the probability that the fault label corresponding to the fault label with the front ranking appears in the power grid is greater, namely, a plurality of fault labels with the front ranking can be selected according to the actual requirement, and the corresponding faults can be determined to be the faults of the current signal data and the fault term data in the corresponding region in the power grid, and can be used as the power grid running state of the region.
According to the technical scheme, the current signal data is subjected to feature extraction through the current feature extraction module comprising the shared learning layer, so that a multi-scale combined architecture for feature extraction of the current signal data can be optimal, and further the extracted current features are optimal multi-scale fusion current features. When semantic feature extraction is carried out on fault term data, the cutoff length of the fault term data is the same as the current feature length, so that granularity difference between the extracted current signal features and text semantic features is reduced, and correlation calculation and cross correlation fusion are conveniently carried out on the fault term data by a cross correlation module. And the output result is obtained by classifying the cross-correlation fused features, and possible faults in the power grid are determined according to the number of fault labels in the output result and the probability of the fault labels corresponding to the fault labels, so that the running state of the power grid is determined, the utilization rate of the acquired data in the running process of the power grid in the running state determination of the power grid is improved, and the accuracy of the running state determination of the power grid is improved.
Example III
Fig. 6 is a schematic structural diagram of a power grid operation state determining device according to a third embodiment of the present invention, where, as shown in fig. 6, the power grid operation state determining device includes a data obtaining module 31, a model processing module 32, and an operation state determining module 33.
The data acquisition module 31 is used for acquiring current signal data and fault term data; the model processing module 32 is used for inputting the current signal data and the fault term data into a pre-trained power grid running state identification model; the power grid running state identification model comprises a cross correlation module for performing characteristic cross correlation processing on current signal data and fault term data; the operation state determining module 33 is configured to determine the operation state of the power grid according to the output result of the power grid operation state identification model.
According to the technical scheme, when the power grid fault is diagnosed through the data collected in the power grid monitoring system, and further the monitoring of the power grid running state is achieved, the current signal data of the electric signal type and the fault term data of the text type collected in the system are input into a pre-trained power grid running state identification model to be processed, the cross-correlation processing is carried out on the characteristics extracted from the current signal data and the fault term data through a cross-correlation module in the power grid running state identification model, the output result of the power grid running state identification model is obtained according to the feature information after the cross-correlation processing, and then the power grid running state is determined according to the output result of the model. The method has the advantages that various power grid operation information acquired in the power grid informatization system can be fully utilized to judge the power grid operation state, the problem that diversified data are difficult to comprehensively represent the same power grid fault due to semantic granularity difference, heterogeneous characteristics and the like is solved, fusion of current data and semantic data in the power grid is realized, the determination of the power grid operation state is realized from the comprehensive direction of multiple dimensions, the utilization rate of the acquired data in the power grid operation state determination in the power grid operation process is improved, the accuracy of the power grid operation state determination is improved, and the power grid fault diagnosis performance is improved.
Optionally, the power grid operation state identification model further includes: the device comprises a semantic feature extraction module, a current feature extraction module and a feature classification module.
Model processing module 32, comprising:
the current characteristic extraction unit is used for inputting the current signal data to the current characteristic extraction module to perform characteristic extraction and determine the current signal characteristics;
the semantic feature extraction unit is used for inputting the fault term data to the semantic feature extraction module for feature extraction and determining text semantic features;
the feature fusion unit is used for inputting the current signal features and the text semantic features into the cross correlation module for cross correlation processing, and determining fusion features;
and the classification output unit is used for inputting the fusion characteristics into the characteristic classification module to perform fault division, and determining the determined fault label probability as an output result of the power grid running state identification model.
Optionally, the current feature extraction module includes a multi-scale coarse grain layer, a first convolution layer, a first shared learning layer, a second convolution layer, a second shared learning layer, a third convolution layer, and a full connection layer; correspondingly, the current characteristic extraction unit is specifically used for:
inputting current signal data into a multi-scale coarse-grained layer, and generating a current information sequence corresponding to each coarse-grained scale factor according to a preset number of coarse-grained scale factors;
Inputting each current information sequence into a first convolution layer for convolution operation, and determining a first output characteristic corresponding to each current information sequence;
inputting each first output characteristic into a first shared learning layer so as to multiply a first output characteristic matrix formed by each first output characteristic with a first shared matrix in the first shared learning layer and determine a first shared output characteristic corresponding to each first output characteristic;
inputting each first shared output characteristic into a second convolution layer to carry out convolution operation, and determining a second output characteristic corresponding to each first shared output characteristic;
inputting each second output characteristic into a second shared learning layer so as to multiply a second output characteristic matrix formed by each second output characteristic with a second shared matrix in the second shared learning layer and determine a second shared output characteristic corresponding to each second output characteristic;
inputting each second shared output characteristic into a third convolution layer to carry out convolution operation, and determining a third output characteristic corresponding to each second shared output characteristic;
inputting each third output characteristic into the full-connection layer for characteristic fusion, and determining current signal characteristics;
the convolution kernels arranged in the first convolution layer, the second convolution layer and the third convolution layer are matched with each coarse grain scale factor.
Optionally, the feature fusion unit is specifically configured to:
inputting the current signal features and the text semantic features into a cross correlation module, and determining the maximum average difference distance between the current signal features and the text semantic features;
and generating fusion features by fusing the current signal features and the text semantic features by minimizing the maximum average difference distance.
Optionally, the operation state determining module 33 is specifically configured to:
determining the number of fault labels contained in an output result and the probability of the fault label corresponding to each fault label;
if the number of the fault labels is smaller than a preset label number threshold value, determining the running state of the power grid according to the corresponding faults of each fault label;
if the number of the fault labels is larger than a preset label number threshold, determining the fault label corresponding to the fault label probability larger than the preset probability threshold as a to-be-determined fault label, sorting the to-be-determined fault labels according to the fault label probability, and determining the running state of the power grid according to the sorting result.
Optionally, the current signal features are consistent with the data length of the text semantic features.
Further, the power grid operation state determining device further includes: model training module for:
before current signal data and fault term data are acquired, acquiring a historical current signal data set, a historical fault term data set and historical grid faults corresponding to the historical current signal data and the historical fault term data in preset historical time;
The method comprises the steps of taking a historical power grid fault as a tag of historical current signal data and historical fault term data, and constructing a training sample set;
and inputting the training sample set into an initial power grid running state identification model, constructing a loss function based on the minimum maximum average difference distance, and training the initial power grid running state identification model until a preset stopping condition is met to obtain the power grid running state identification model.
The power grid running state determining device provided by the embodiment of the invention can execute the power grid running state determining method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the executing method.
Example IV
Fig. 7 is a schematic structural diagram of a power grid operation state determining device according to a fourth embodiment of the present invention. The grid operating condition determining device 40 may be an electronic device intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.
As shown in fig. 7, the grid operation state determination device 40 includes at least one processor 41, and a memory, such as a Read Only Memory (ROM) 42, a Random Access Memory (RAM) 43, etc., communicatively connected to the at least one processor 41, in which the memory stores a computer program executable by the at least one processor, and the processor 41 may perform various appropriate actions and processes according to the computer program stored in the Read Only Memory (ROM) 42 or the computer program loaded from the storage unit 48 into the Random Access Memory (RAM) 43. In the RAM 43, various programs and data required for the operation of the grid operation state determination device 40 may also be stored. The processor 41, the ROM 42 and the RAM 43 are connected to each other via a bus 44. An input/output (I/O) interface 45 is also connected to bus 44.
The various components in the grid operating condition determining device 40 are connected to the I/O interface 45, including: an input unit 46 such as a keyboard, a mouse, etc.; an output unit 47 such as various types of displays, speakers, and the like; a storage unit 48 such as a magnetic disk, an optical disk, or the like; and a communication unit 49 such as a network card, modem, wireless communication transceiver, etc. The communication unit 49 allows the grid operation status determination device 40 to exchange information/data with other devices via a computer network such as the internet and/or various telecommunication networks.
The processor 41 may be various general and/or special purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 41 performs the various methods and processes described above, such as the grid operating condition determination method.
In some embodiments, the grid operating condition determination method may be implemented as a computer program tangibly embodied on a computer-readable storage medium, such as the storage unit 48. In some embodiments, part or all of the computer program may be loaded and/or installed onto the grid operation status determination device 40 via the ROM 42 and/or the communication unit 49. When the computer program is loaded into RAM 43 and executed by processor 41, one or more steps of the grid operation status determination method described above may be performed. Alternatively, in other embodiments, the processor 41 may be configured to perform the grid operating state determination method by any other suitable means (e.g., by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
A computer program for carrying out methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.
The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.
The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.

Claims (10)

1. A method for determining an operating state of a power grid, comprising:
acquiring current signal data and fault term data;
inputting the current signal data and the fault term data into a pre-trained power grid running state identification model; the power grid running state identification model comprises a cross correlation module for performing characteristic cross correlation processing on the current signal data and the fault term data;
And determining the running state of the power grid according to the output result of the power grid running state identification model.
2. The method of claim 1, wherein the grid operating condition identification model further comprises: the device comprises a semantic feature extraction module, a current feature extraction module and a feature classification module;
the step of inputting the current signal data and the fault term data into a pre-trained power grid running state identification model comprises the following steps:
inputting the current signal data to the current feature extraction module for feature extraction, and determining current signal features;
inputting the fault term data to the semantic feature extraction module for feature extraction, and determining text semantic features;
inputting the current signal characteristics and the text semantic characteristics into the cross correlation module to perform cross correlation processing, and determining fusion characteristics;
and inputting the fusion characteristics to the characteristic classification module for fault division, and determining the determined fault label probability as an output result of the power grid running state identification model.
3. The method of claim 2, wherein the current feature extraction module comprises a multi-scale coarse grain layer, a first convolution layer, a first shared learning layer, a second convolution layer, a second shared learning layer, a third convolution layer, and a fully connected layer;
The step of inputting the current signal data to the current feature extraction module for feature extraction, and determining the current signal features comprises the following steps:
inputting the current signal data into the multi-scale coarse-grained layer, and generating a current information sequence corresponding to each coarse-grained scale factor according to a preset number of coarse-grained scale factors;
inputting each current information sequence into the first convolution layer to carry out convolution operation, and determining a first output characteristic corresponding to each current information sequence;
inputting each first output characteristic to the first shared learning layer so as to multiply a first output characteristic matrix formed by each first output characteristic with a first shared matrix in the first shared learning layer to determine a first shared output characteristic corresponding to each first output characteristic;
inputting each first shared output characteristic into a second convolution layer to carry out convolution operation, and determining a second output characteristic corresponding to each first shared output characteristic;
inputting each second output characteristic to the second shared learning layer so as to multiply a second output characteristic matrix formed by each second output characteristic with a second shared matrix in the second shared learning layer, and determining a second shared output characteristic corresponding to each second output characteristic;
Inputting each second shared output characteristic into a third convolution layer to carry out convolution operation, and determining a third output characteristic corresponding to each second shared output characteristic;
inputting each third output characteristic to the full-connection layer for characteristic fusion, and determining current signal characteristics;
and the convolution kernels arranged in the first convolution layer, the second convolution layer and the third convolution layer are matched with the coarse grain scale factors.
4. The method of claim 2, wherein the inputting the current signal feature and the text semantic feature to the cross-correlation module for cross-correlation processing, determining a fusion feature, comprises:
inputting the current signal features and the text semantic features to the cross correlation module, and determining the maximum average difference distance between the current signal features and the text semantic features;
and fusing the current signal characteristics and the text semantic characteristics by minimizing the maximum average difference distance to generate fused characteristics.
5. The method according to claim 2, wherein determining the grid operation state according to the output result of the grid operation state identification model comprises:
Determining the number of fault labels contained in the output result and the probability of the fault label corresponding to each fault label;
if the number of the fault labels is smaller than a preset label number threshold value, determining the running state of the power grid according to the corresponding faults of each fault label;
if the number of the fault labels is larger than the preset label number threshold, determining the fault label corresponding to the fault label probability larger than the preset probability threshold as a to-be-determined fault label, sorting the to-be-determined fault labels according to the fault label probability, and determining the running state of the power grid according to the sorting result.
6. The method of any of claims 2-5, wherein the current signal features are consistent with data lengths of the text semantic features.
7. The method of any of claims 1-5, further comprising, prior to the obtaining the current signal data and the fault term data:
acquiring a historical current signal data set, a historical fault term data set and historical grid faults corresponding to each historical current signal data and historical fault term data in preset historical time;
the historical power grid faults are used as labels of the historical current signal data and the historical fault term data, and a training sample set is constructed;
And inputting the training sample set into an initial power grid running state identification model, constructing a loss function based on the minimum average difference distance, and training the initial power grid running state identification model until a preset stopping condition is met to obtain the power grid running state identification model.
8. An electrical grid operating condition determining apparatus, comprising:
the data acquisition module is used for acquiring current signal data and fault term data;
the model processing module is used for inputting the current signal data and the fault term data into a pre-trained power grid running state identification model; the power grid running state identification model comprises a cross correlation module for performing characteristic cross correlation processing on the current signal data and the fault term data;
and the running state determining module is used for determining the running state of the power grid according to the output result of the power grid running state identification model.
9. A power grid operation state determination apparatus, characterized by comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein, the liquid crystal display device comprises a liquid crystal display device,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the grid operation state determination method of any one of claims 1-7.
10. A storage medium containing computer executable instructions which, when executed by a computer processor, are for performing the grid operating condition determination method as claimed in any one of claims 1 to 7.
CN202310721388.3A 2023-06-16 2023-06-16 Power grid running state determining method, device, equipment and storage medium Pending CN116914921A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310721388.3A CN116914921A (en) 2023-06-16 2023-06-16 Power grid running state determining method, device, equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310721388.3A CN116914921A (en) 2023-06-16 2023-06-16 Power grid running state determining method, device, equipment and storage medium

Publications (1)

Publication Number Publication Date
CN116914921A true CN116914921A (en) 2023-10-20

Family

ID=88350154

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310721388.3A Pending CN116914921A (en) 2023-06-16 2023-06-16 Power grid running state determining method, device, equipment and storage medium

Country Status (1)

Country Link
CN (1) CN116914921A (en)

Similar Documents

Publication Publication Date Title
CN108228325B (en) Application management method and device, electronic equipment and computer storage medium
EP3961476A1 (en) Entity linking method and apparatus, electronic device and storage medium
US11809505B2 (en) Method for pushing information, electronic device
WO2022010575A1 (en) Automatic recognition of entities related to cloud incidents
CN115294397A (en) Classification task post-processing method, device, equipment and storage medium
CN117041017A (en) Intelligent operation and maintenance management method and system for data center
CN115794578A (en) Data management method, device, equipment and medium for power system
CN114090601B (en) Data screening method, device, equipment and storage medium
CN114037059A (en) Pre-training model, model generation method, data processing method and data processing device
CN117234844A (en) Cloud server abnormality management method and device, computer equipment and storage medium
CN116755974A (en) Cloud computing platform operation and maintenance method and device, electronic equipment and storage medium
CN115600607A (en) Log detection method and device, electronic equipment and medium
CN116361567A (en) Data processing method and system applied to cloud office
CN116467606A (en) Determination method, device, equipment and medium of decision suggestion information
CN116914921A (en) Power grid running state determining method, device, equipment and storage medium
CN114610953A (en) Data classification method, device, equipment and storage medium
CN113051911B (en) Method, apparatus, device, medium and program product for extracting sensitive words
CN115603955A (en) Abnormal access object identification method, device, equipment and medium
CN114328123A (en) Abnormality determination method, training method, device, electronic device, and storage medium
CN114120180A (en) Method, device, equipment and medium for generating time sequence nomination
CN117574146B (en) Text classification labeling method, device, electronic equipment and storage medium
CN112329427B (en) Method and device for acquiring short message samples
CN117649115A (en) Risk assessment method and device, electronic equipment and storage medium
CN116720186A (en) Malicious code identification method and device, electronic equipment and storage medium
CN116431809A (en) Text labeling method, device and storage medium based on bank customer service scene

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