CN114066262A - Method, system and device for estimating cause-tracing reasoning of abnormal indexes after power grid dispatching and storage medium - Google Patents

Method, system and device for estimating cause-tracing reasoning of abnormal indexes after power grid dispatching and storage medium Download PDF

Info

Publication number
CN114066262A
CN114066262A CN202111367112.7A CN202111367112A CN114066262A CN 114066262 A CN114066262 A CN 114066262A CN 202111367112 A CN202111367112 A CN 202111367112A CN 114066262 A CN114066262 A CN 114066262A
Authority
CN
China
Prior art keywords
abnormal
factors
index
reasoning
power grid
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
CN202111367112.7A
Other languages
Chinese (zh)
Inventor
闫朝阳
仇晨光
熊浩
张振华
崔占飞
戴上
赵玉林
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.)
State Grid Jiangsu Electric Power Co Ltd
Original Assignee
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 State Grid Jiangsu Electric Power Co Ltd filed Critical State Grid Jiangsu Electric Power Co Ltd
Priority to CN202111367112.7A priority Critical patent/CN114066262A/en
Publication of CN114066262A publication Critical patent/CN114066262A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • 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
    • G06F18/24155—Bayesian classification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0635—Risk analysis of enterprise or organisation activities
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply

Landscapes

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

Abstract

The invention discloses a method, a system, a device and a storage medium for estimating cause-tracing reasoning of abnormal indexes after power grid dispatching, wherein the method comprises the following steps: acquiring time sequence data in the running process of a power grid, and extracting abnormal indexes and various associated factors according to the time sequence data; calculating the association degree of the abnormal index and each association factor to obtain a strong correlation factor; calculating the causal relationship between the abnormal indexes and each associated factor to obtain strong causal factors; constructing a Bayesian network based on the strong correlation factors, the strong causal factors and the abnormal indexes; and (4) parameter learning is carried out on the Bayesian network, and the cause tracing reasoning of the abnormal index is carried out on the basis of the Bayesian network after the parameter learning. The method can solve the key technical problem of analysis of the leading factors of the abnormal indexes.

Description

Method, system and device for estimating cause-tracing reasoning of abnormal indexes after power grid dispatching and storage medium
Technical Field
The invention relates to a method, a system and a device for reasoning on evaluation abnormal indexes after power grid dispatching and a storage medium, and belongs to the technical field of power system automation.
Background
The power grid operation index is an important carrier for scientifically evaluating the safe and efficient operation of the power grid and guiding the optimal operation of the power grid, and is a technical means for effectively improving the lean level of the power grid operation. At present, more researches on power grid operation evaluation indexes are carried out, but for the change rule of the power grid operation indexes, the positioning of abnormal indexes and reason analysis, an automatic and intelligent evaluation means is lacked, operators are often required to carry out manual analysis by combining a large amount of historical data and manual experience, the workload is large, the efficiency is not high, and the lean operation requirements of the power grid cannot be completely met. The method has the advantages that mass data are generated in the operation process of the power grid, evaluation indexes are only considered from different sides, correlation analysis is lacked, and the generation reason of abnormal indexes is difficult to accurately position.
The power system has the characteristics of real-time balance of power generation and power utilization, no stop of real-time operation, diffusibility in accidents and the like, and mass data, rapid growth and various types of data are generated in the operation process of the power system.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, provides a reasoning method, a system, a device and a storage medium for evaluating the abnormal index traceability after power grid dispatching, can find out the main reason of the index abnormality by constructing a Bayesian network, and has strong practicability and flexibility.
In order to achieve the purpose, the invention is realized by adopting the following technical scheme:
in a first aspect, the invention provides a method for reasoning and estimating a cause of an abnormal index after power grid dispatching, which comprises the following steps:
acquiring time sequence data in the running process of a power grid, and extracting abnormal indexes and various associated factors according to the time sequence data;
calculating the association degree of the abnormal index and each association factor to obtain a strong correlation factor;
calculating the causal relationship between the abnormal indexes and each associated factor to obtain strong causal factors;
constructing a Bayesian network based on the strong correlation factors, the strong causal factors and the abnormal indexes;
and (4) parameter learning is carried out on the Bayesian network, and the cause tracing reasoning of the abnormal index is carried out on the basis of the Bayesian network after the parameter learning.
And the abnormal index is an index of the time series data deviating from the normal time series distribution curve.
Optionally, the time series data includes power grid equipment parameter data, power grid operation parameter data, power grid node power generation and load parameter data, and environmental parameter data.
Optionally, the calculating the association degree between the abnormal index and each association factor, and the obtaining the strong association factor includes:
constructing a data set of the abnormal indexes and the relevant factors:
Figure BDA0003361061990000021
in the formula, x0Time series data corresponding to the abnormality index, xi(i-1, 2, … n) is time series data corresponding to each related factor，xn(N) nth time series data for the nth correlation factor;
finding an anomaly index x from a data set0The difference matrix sequence with the ith correlation factor:
Δ0i(l)＝|x0(l)-xi(l)|，l＝1,2,…,N；i＝1,2,…,n
solving the abnormal index x according to the difference matrix sequence0Correlation coefficient with i-th correlation factor:
Figure BDA0003361061990000031
in the formula,. DELTA.minAnd ΔmaxRespectively a minimum value and a maximum value in the difference matrix sequence, wherein rho is a resolution coefficient; obtaining an abnormal index x according to the correlation coefficient0Degree of association with i-th association factor:
Figure BDA0003361061990000032
correlation degree r is taken0iCorrelation factor x greater than or equal to preset correlation thresholdiAs an abnormality index x0Is strongly correlated.
Optionally, the calculating a causal relationship between the abnormal index and each of the associated factors, and acquiring a strong causal factor includes:
and (3) constructing a linear autoregressive model of the abnormal index and the time series data corresponding to any correlation factor:
Figure BDA0003361061990000033
in the formula, XtTime series data corresponding to the abnormality index, YtTime sequence data corresponding to any correlation factor; a is1jAnd d1jIs a fitting coefficient of an autoregressive model, epsilon1tAnd η1tIs a residual error, sigma1And Γ1Is a residual epsilon1tAnd η1tThe variance of (a);
performing joint regression on the linear autoregressive model to construct a regression model:
Figure BDA0003361061990000041
in the formula, a2j、b2j、c2jAnd d2jIs the fitting coefficient of the regression model, ε2tAnd η2tFor time-independent prediction errors, sigma2And Γ2To predict the error e2tAnd η2tThe variance of (a);
calculating YtTo XtGlangager cause and effect of (a):
Figure BDA0003361061990000042
if FY→XIf it is greater than the predetermined threshold value, YtIs XtThe glange factor of (a), there is a causal relationship;
and taking the correlation factors with the causal relationship as strong causal factors of the abnormal indexes.
Optionally, the constructing a bayesian network based on the strong correlation factor, the strong causal factor and the abnormal index includes:
constructing a node at the bottom layer of the Bayesian network through the abnormal indexes, and recording the node as a leaf node;
constructing nodes above the leaf nodes through strong correlation factors and strong causal factors, and recording the nodes as intermediate nodes;
the node positioned at the topmost layer of the middle node is marked as a root node;
the nodes are connected through directed edges, and the directed edges represent the relationship among the nodes.
Optionally, the parameter learning for the bayesian network includes: discretizing node data with continuous characteristics in the Bayesian network by using an equal division method; and learning by using discretized node data and adopting a maximum likelihood estimation method.
Optionally, the performing of the cause-tracing reasoning on the abnormal index based on the bayesian network after parameter learning includes:
definition of
Figure BDA0003361061990000043
And TqIs a root node xi(i ═ 1,2, …, m), intermediate node yj(j ═ 1,2 …, n) and the risk status of leaf node T, where ai＝1,2,…,ki；bj＝1,2,…,kj；q＝1,2,…,kq；ki、kj、kqRespectively, the risk state number;
compute root node xiThe risk state is
Figure BDA0003361061990000051
For leaf node T the risk state is TqProbability importance of (2):
Figure BDA0003361061990000052
in the formula (I), the compound is shown in the specification,
Figure BDA0003361061990000053
in order to be the importance of the probability,
Figure BDA0003361061990000054
is a root node xiThe risk state is
Figure BDA0003361061990000055
The T risk state of the time leaf node is TqP (T ═ T)q|xi1) is root node xiThe leaf node T risk state is T in the initial stateqThe probability of (d);
calculating the key importance according to the probability importance:
Figure BDA0003361061990000056
the average of the key importance is found:
Figure BDA0003361061990000057
sorting according to the average value of the key importance degrees, and obtaining the root node x with larger valueiThe corresponding correlation factors are key risk factors causing the generation of the abnormal indexes and are output as the reason-tracing reasoning result.
In a second aspect, the invention provides a system for reasoning and reasoning reasons for evaluating abnormal indexes after power grid dispatching, which is characterized by comprising:
the data extraction module is used for acquiring time sequence data in the running process of the power grid and extracting abnormal indexes and various associated factors according to the time sequence data;
the strong correlation factor acquisition module is used for calculating the correlation degree of the abnormal index and each correlation factor to acquire the strong correlation factor;
the strong causal relationship acquisition module is used for calculating the causal relationship between the abnormal index and each associated factor to acquire a strong causal factor;
the Bayesian network construction module is used for constructing a Bayesian network based on strong correlation factors, strong causal factors and abnormal indexes;
and the traceability reasoning module is used for parameter learning of the Bayesian network and traceability reasoning of the abnormal indexes based on the Bayesian network after the parameter learning.
And the abnormal index is an index of the time series data deviating from the normal time series distribution curve.
In a third aspect, the invention provides a device for estimating the cause-tracing reasoning of abnormal indexes after power grid dispatching, which is characterized by comprising a processor and a storage medium;
the storage medium is used for storing instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method according to any of the above.
In a fourth aspect, the invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, controls an apparatus in which the storage medium is located to perform the steps of any of the methods described above.
Compared with the prior art, the invention has the following beneficial effects:
according to the method, the system, the device and the storage medium for reasoning the abnormal index after power grid scheduling, strong correlation factors and strong cause-and-effect factors are obtained by analyzing the relevance and cause-and-effect relationship of the abnormal index, a Bayesian network is constructed based on the strong correlation factors and the strong cause-and-effect factors, the risk factors are identified through key importance indexes by applying a Bayesian network reasoning technology, and therefore the main reasons of the abnormal index are found, and the method has strong practicability and flexibility.
Drawings
Fig. 1 is a flowchart of a method for reasoning a cause of an abnormal evaluation indicator after power grid scheduling according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for obtaining strong correlation factors according to an embodiment of the present invention;
fig. 3 is a flowchart of a method for obtaining strong causal factors according to an embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
The first embodiment is as follows:
as shown in fig. 1, an embodiment of the present invention provides a method for reasoning and reasoning a cause of an abnormal evaluation indicator after power grid scheduling, including the following steps:
(1) acquiring time sequence data in the running process of a power grid, and extracting abnormal indexes and various associated factors according to the time sequence data;
the abnormal index is an index of the deviation of time series data from a normal time series distribution curve.
The time sequence data comprises power grid equipment parameter data, power grid operation parameter data, power grid node power generation and load parameter data and environment parameter data.
(2) Calculating the association degree of the abnormal index and each association factor to obtain a strong correlation factor;
as shown in fig. 2:
2.1, constructing a data set of the abnormal indexes and the relevant factors:
Figure BDA0003361061990000071
in the formula, x0Time series data corresponding to the abnormality index, xi(i 1,2, … n) is time series data corresponding to each of the related factors, xn(N) nth time series data for the nth correlation factor;
2.2, obtaining the abnormal index x according to the data set0The difference matrix sequence with the ith correlation factor:
Δ0i(l)＝|x0(l)-xi(l)|，l＝1,2,…,N；i＝1,2,…,n
2.3, solving the abnormal index x according to the difference matrix sequence0Correlation coefficient with i-th correlation factor:
Figure BDA0003361061990000081
in the formula,. DELTA.minAnd ΔmaxRespectively a minimum value and a maximum value in the difference matrix sequence, wherein rho is a resolution coefficient;
2.4, obtaining the abnormal index x according to the correlation coefficient0Degree of association with i-th association factor:
Figure BDA0003361061990000082
2.5, taking the degree of association r0iThe correlation factor x is greater than or equal to a preset correlation threshold (generally 0.8)iAs an abnormality index x0Is strongly correlated.
(3) Calculating the causal relationship between the abnormal indexes and each associated factor to obtain strong causal factors;
as shown in fig. 3:
3.1, constructing a linear autoregressive model of the abnormal index and the time sequence data corresponding to any correlation factor:
Figure BDA0003361061990000083
in the formula, XtTime series data corresponding to the abnormality index, YtTime sequence data corresponding to any correlation factor; a is1jAnd d1jIs a fitting coefficient of an autoregressive model, epsilon1tAnd η1tIs a residual error, sigma1And Γ1Is a residual epsilon1tAnd η1tThe variance of (a);
3.2, carrying out joint regression on the linear autoregressive model to construct a regression model:
Figure BDA0003361061990000091
in the formula, a2j、b2j、c2jAnd d2jIs the fitting coefficient of the regression model, ε2tAnd η2tFor time-independent prediction errors, sigma2And Γ2To predict the error e2tAnd η2tThe variance of (a);
3.3, calculating any correlation factor YtFor abnormal index XtGlangager cause and effect of (a):
Figure BDA0003361061990000092
if FY→XIf it is greater than the predetermined threshold value, YtIs XtThe glange factor of (a), there is a causal relationship;
and 3.4, taking the correlation factors of the causal relationship as strong causal factors of the abnormal indexes.
(4) Constructing a Bayesian network based on the strong correlation factors, the strong causal factors and the abnormal indexes;
4.1, constructing a node at the bottommost layer of the Bayesian network through the abnormal indexes, and recording the node as a leaf node;
4.2, constructing nodes above the leaf nodes through strong correlation factors and strong causal factors, and marking as intermediate nodes;
4.3, marking the node positioned at the topmost layer of the middle node as a root node;
and 4.4, connecting the nodes through directed edges, wherein the directed edges represent the relationship between the nodes.
(5) And (4) parameter learning is carried out on the Bayesian network, and the cause tracing reasoning of the abnormal index is carried out on the basis of the Bayesian network after the parameter learning.
The parameter learning of the bayesian network comprises: discretizing node data with continuous characteristics in the Bayesian network by using an equal division method; and learning by using discretized node data and adopting a maximum likelihood estimation method.
Performing tracing reasoning of the abnormal indexes based on the Bayesian network after parameter learning comprises the following steps:
definition of
Figure BDA0003361061990000101
And TqIs a root node xi(i ═ 1,2, …, m), intermediate node yj(j ═ 1,2 …, n) and the risk status of leaf node T, where ai＝1,2,…,ki；bj＝1,2,…,kj；q＝1,2,…,kq；ki、kj、kqRespectively, the risk state number;
compute root node xiThe risk state is
Figure BDA0003361061990000102
For leaf node T the risk state is TqProbability importance of (2):
Figure BDA0003361061990000103
in the formula (I), the compound is shown in the specification,
Figure BDA0003361061990000104
in order to be the importance of the probability,
Figure BDA0003361061990000105
is a root node xiThe risk state is
Figure BDA0003361061990000106
The T risk state of the time leaf node is TqP (T ═ T)q|xi1) is root node xiThe leaf node T risk state is T in the initial stateqThe probability of (d);
calculating the key importance according to the probability importance:
Figure BDA0003361061990000107
the average of the key importance is found:
Figure BDA0003361061990000108
sorting according to the average value of the key importance degrees, and obtaining the root node x with larger valueiThe corresponding correlation factors are key risk factors causing the generation of the abnormal indexes and are output as the reason-tracing reasoning result.
In the embodiment, a grey correlation method is used for carrying out correlation analysis on the indexes and the correlation factors, a Glange causal analysis method is used for carrying out causal analysis on the indexes and the strong correlation factors, the indexes and the strong correlation factors can be effectively screened by combining the indexes and the strong correlation factors, the weak correlation factors are eliminated, and a foundation is provided for establishing a Bayesian structure.
The method establishes a Bayesian network to carry out relevance graph representation on the indexes and the relevance factors, and identifies risk factors through key importance indexes by using a Bayesian network reasoning technology, so that the main reasons of index abnormality are found, and the method has strong practicability and flexibility.
Example two:
the embodiment of the invention provides a system for estimating abnormal index traceability reasoning after power grid dispatching, which comprises:
the data extraction module is used for acquiring time sequence data in the running process of the power grid and extracting abnormal indexes and various associated factors according to the time sequence data;
the strong correlation factor acquisition module is used for calculating the correlation degree of the abnormal index and each correlation factor to acquire the strong correlation factor;
the strong causal relationship acquisition module is used for calculating the causal relationship between the abnormal index and each associated factor to acquire a strong causal factor;
the Bayesian network construction module is used for constructing a Bayesian network based on strong correlation factors, strong causal factors and abnormal indexes;
and the traceability reasoning module is used for parameter learning of the Bayesian network and traceability reasoning of the abnormal indexes based on the Bayesian network after the parameter learning.
Wherein, the abnormal index is an index of the time sequence data deviating from the normal time sequence distribution curve.
Example three:
based on the method for reasoning the traceability of the abnormal indexes in power grid dispatching evaluation, the embodiment of the invention provides a device for reasoning the traceability of the abnormal indexes after power grid dispatching evaluation, which comprises a processor and a storage medium;
a storage medium to store instructions;
the processor is configured to operate in accordance with instructions to perform steps in accordance with the above-described method.
Example four:
based on the method for reasoning the tracing cause of the abnormal index of the power grid dispatching evaluation, the embodiment of the invention provides a computer readable storage medium, wherein a computer program is stored on the storage medium, and the method is characterized in that when the program is run by a processor, the program controls equipment where the storage medium is located to execute the steps of the method.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.

Claims (10)

1. A reason-tracing reasoning method for evaluating abnormal indexes after power grid dispatching is characterized by comprising the following steps:
acquiring time sequence data in the running process of a power grid, and extracting abnormal indexes and various associated factors according to the time sequence data;
calculating the association degree of the abnormal index and each association factor to obtain a strong correlation factor;
calculating the causal relationship between the abnormal indexes and each associated factor to obtain strong causal factors;
constructing a Bayesian network based on the strong correlation factors, the strong causal factors and the abnormal indexes;
and (4) parameter learning is carried out on the Bayesian network, and the cause tracing reasoning of the abnormal index is carried out on the basis of the Bayesian network after the parameter learning.
And the abnormal index is an index of the time series data deviating from the normal time series distribution curve.
2. The method for reasoning on the evaluation anomaly index tracing reason after power grid dispatching according to claim 1, wherein the time sequence data comprises power grid equipment parameter data, power grid operation parameter data, power grid node power generation and load parameter data and environment parameter data.
3. The method for reasoning and tracing to the cause of the evaluation abnormal index after the power grid dispatching according to claim 1, wherein the calculating the degree of association between the abnormal index and each associated factor and the obtaining of the strong associated factor comprises:
constructing a data set of the abnormal indexes and the relevant factors:
Figure FDA0003361061980000011
in the formula, x0Time series data corresponding to the abnormality index, xi(i 1,2, … n) is time series data corresponding to each of the related factors, xn(N) nth time series data for the nth correlation factor;
finding an anomaly index x from a data set0The difference matrix sequence with the ith correlation factor:
Δ0i(l)＝|x0(l)-xi(l)|，l＝1,2,…,N；i＝1,2,…,n
solving the abnormal index x according to the difference matrix sequence0Correlation coefficient with i-th correlation factor:
Figure FDA0003361061980000021
in the formula,. DELTA.minAnd ΔmaxRespectively a minimum value and a maximum value in the difference matrix sequence, wherein rho is a resolution coefficient;
obtaining an abnormal index x according to the correlation coefficient0Degree of association with i-th association factor:
Figure FDA0003361061980000022
correlation degree r is taken0iCorrelation factor x greater than or equal to preset correlation thresholdiAs an abnormality index x0Is strongly correlated.
4. The method for reasoning on the evaluation anomaly index after power grid scheduling according to claim 1, wherein the calculating of the causal relationship between the anomaly index and each correlation factor and the obtaining of the strong causal factor comprise:
and (3) constructing a linear autoregressive model of the abnormal index and the time series data corresponding to any correlation factor:
Figure FDA0003361061980000023
in the formula, XtTime series data corresponding to the abnormality index, YtTime sequence data corresponding to any correlation factor; a is1jAnd d1jIs a fitting coefficient of an autoregressive model, epsilon1tAnd η1tIs a residual error, sigma1And Γ1Is a residual epsilon1tAnd η1tThe variance of (a);
performing joint regression on the linear autoregressive model to construct a regression model:
Figure FDA0003361061980000031
in the formula, a2j、b2j、c2jAnd d2jIs the fitting coefficient of the regression model, ε2tAnd η2tFor time-independent prediction errors, sigma2And Γ2To predict the error e2tAnd η2tThe variance of (a);
calculating YtTo XtGlangager cause and effect of (a):
Figure FDA0003361061980000032
if FY→XIf it is greater than the predetermined threshold value, YtIs XtThe glange factor of (a), there is a causal relationship;
and taking the correlation factors with the causal relationship as strong causal factors of the abnormal indexes.
5. The method for reasoning on the evaluation anomaly indicators after power grid dispatching according to claim 1, wherein the constructing the Bayesian network based on the strong correlation factors, the strong causal factors and the anomaly indicators comprises:
constructing a node at the bottom layer of the Bayesian network through the abnormal indexes, and recording the node as a leaf node;
constructing nodes above the leaf nodes through strong correlation factors and strong causal factors, and recording the nodes as intermediate nodes;
the node positioned at the topmost layer of the middle node is marked as a root node;
the nodes are connected through directed edges, and the directed edges represent the relationship among the nodes.
6. The method for reasoning on the evaluation anomaly index after power grid dispatching according to claim 1, wherein the parameter learning of the Bayesian network comprises: discretizing node data with continuous characteristics in the Bayesian network by using an equal division method; and learning by using discretized node data and adopting a maximum likelihood estimation method.
7. The method for reasoning the estimated abnormal index on the basis of the power grid after dispatching according to claim 1, wherein the reasoning the abnormal index on the basis of the Bayesian network after parameter learning comprises the following steps:
definition of
Figure FDA0003361061980000041
And TqIs a root node xi(i ═ 1,2, …, m), intermediate node yj(j ═ 1,2 …, n) and the risk status of leaf node T, where ai＝1,2,…,ki；bj＝1,2,…,kj；q＝1,2,…,kq；ki、kj、kqRespectively, the risk state number;
compute root node xiThe risk state is
Figure FDA0003361061980000042
For leaf node T the risk state is TqProbability importance of (2):
Figure FDA0003361061980000043
in the formula (I), the compound is shown in the specification,
Figure FDA0003361061980000044
in order to be the importance of the probability,
Figure FDA0003361061980000045
is a root node xiThe risk state is
Figure FDA0003361061980000046
The T risk state of the time leaf node is TqP (T ═ T)q|xi1) is root node xiThe leaf node T risk state is T in the initial stateqThe probability of (d);
calculating the key importance according to the probability importance:
Figure FDA0003361061980000047
the average of the key importance is found:
Figure FDA0003361061980000048
sorting according to the average value of the key importance degrees, and obtaining the root node x with larger valueiThe corresponding correlation factors are key risk factors causing the generation of the abnormal indexes and are output as the reason-tracing reasoning result.
8. A traceability reasoning system for evaluating abnormal indexes after power grid dispatching is characterized by comprising:
the data extraction module is used for acquiring time sequence data in the running process of the power grid and extracting abnormal indexes and various associated factors according to the time sequence data;
the strong correlation factor acquisition module is used for calculating the correlation degree of the abnormal index and each correlation factor to acquire the strong correlation factor;
the strong causal relationship acquisition module is used for calculating the causal relationship between the abnormal index and each associated factor to acquire a strong causal factor;
the Bayesian network construction module is used for constructing a Bayesian network based on strong correlation factors, strong causal factors and abnormal indexes;
and the traceability reasoning module is used for parameter learning of the Bayesian network and traceability reasoning of the abnormal indexes based on the Bayesian network after the parameter learning.
And the abnormal index is an index of the time series data deviating from the normal time series distribution curve.
9. A reason-tracing reasoning device for evaluating abnormal indexes after power grid dispatching is characterized by comprising a processor and a storage medium;
the storage medium is used for storing instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method according to any one of claims 1 to 7.
10. Computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, controls an apparatus of the storage medium to carry out the steps of the method according to any one of claims 1 to 7.
CN202111367112.7A 2021-11-18 2021-11-18 Method, system and device for estimating cause-tracing reasoning of abnormal indexes after power grid dispatching and storage medium Pending CN114066262A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111367112.7A CN114066262A (en) 2021-11-18 2021-11-18 Method, system and device for estimating cause-tracing reasoning of abnormal indexes after power grid dispatching and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111367112.7A CN114066262A (en) 2021-11-18 2021-11-18 Method, system and device for estimating cause-tracing reasoning of abnormal indexes after power grid dispatching and storage medium

Publications (1)

Publication Number Publication Date
CN114066262A true CN114066262A (en) 2022-02-18

Family

ID=80278956

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111367112.7A Pending CN114066262A (en) 2021-11-18 2021-11-18 Method, system and device for estimating cause-tracing reasoning of abnormal indexes after power grid dispatching and storage medium

Country Status (1)

Country Link
CN (1) CN114066262A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114553675A (en) * 2022-03-24 2022-05-27 中国联合网络通信集团有限公司 Fault network element processing method, device and storage medium
CN114693106A (en) * 2022-03-28 2022-07-01 国网山东省电力公司信息通信公司 Cloud edge mode-based data model construction method for information thing networking unification
CN114780619A (en) * 2022-06-07 2022-07-22 国网浙江省电力有限公司金华供电公司 Abnormity early warning method for automatic engineering audit data
CN115080909A (en) * 2022-07-15 2022-09-20 深圳市城市交通规划设计研究中心股份有限公司 Analysis method for influencing data of internet of things sensing equipment, electronic equipment and storage medium
CN116821725A (en) * 2023-03-28 2023-09-29 北京火山引擎科技有限公司 Data processing method, device and storage medium

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114553675A (en) * 2022-03-24 2022-05-27 中国联合网络通信集团有限公司 Fault network element processing method, device and storage medium
CN114553675B (en) * 2022-03-24 2023-05-09 中国联合网络通信集团有限公司 Fault network element processing method, device and storage medium
CN114693106A (en) * 2022-03-28 2022-07-01 国网山东省电力公司信息通信公司 Cloud edge mode-based data model construction method for information thing networking unification
CN114780619A (en) * 2022-06-07 2022-07-22 国网浙江省电力有限公司金华供电公司 Abnormity early warning method for automatic engineering audit data
CN115080909A (en) * 2022-07-15 2022-09-20 深圳市城市交通规划设计研究中心股份有限公司 Analysis method for influencing data of internet of things sensing equipment, electronic equipment and storage medium
CN115080909B (en) * 2022-07-15 2022-11-25 深圳市城市交通规划设计研究中心股份有限公司 Analysis method for influencing data of internet of things sensing equipment, electronic equipment and storage medium
CN116821725A (en) * 2023-03-28 2023-09-29 北京火山引擎科技有限公司 Data processing method, device and storage medium

Similar Documents

Publication Publication Date Title
CN114066262A (en) Method, system and device for estimating cause-tracing reasoning of abnormal indexes after power grid dispatching and storage medium
CN109587713B (en) Network index prediction method and device based on ARIMA model and storage medium
CN103310051A (en) Method for predicting failure rate of vehicle-mounted information terminal in entire life cycle
CN114035468B (en) Method and system for predictively monitoring overhaul flow of fan based on XGBoost algorithm
Azimi et al. Applying basic control theory principles to project control: Case study of off-site construction shops
CN113868953B (en) Multi-unit operation optimization method, device and system in industrial system and storage medium
CN111461450A (en) Hydroelectric generating set degradation degree prediction method based on EEMD and L STM
CN115438726A (en) Device life and fault type prediction method and system based on digital twin technology
CN114970926A (en) Model training method, enterprise operation risk prediction method and device
CN116467674A (en) Intelligent fault processing fusion updating system and method for power distribution network
CN113190457A (en) Software reliability evaluation method for networked system
CN113742993B (en) Method, device, equipment and storage medium for predicting life loss of dry-type transformer
CN115238573A (en) Hydroelectric generating set performance degradation trend prediction method and system considering working condition parameters
Xu et al. An adaptive gamma process based model for residual useful life prediction
Wang et al. Degradation prediction method by use of autoregressive algorithm
CN112700050A (en) Method and system for predicting ultra-short-term 1 st point power of photovoltaic power station
Chua et al. A surrogate model to predict production performance in digital twin-based smart manufacturing
CN109635008B (en) Equipment fault detection method based on machine learning
US20140052425A1 (en) Method and apparatus for evaluating a model of an industrial plant process
Khalid et al. Early cost estimation of software reworks using fuzzy requirement-based model
CN114818460A (en) Laboratory equipment residual service life prediction method based on automatic machine learning
CN113139332A (en) Automatic model construction method, device and equipment
CN114781473A (en) Method, device and equipment for predicting state of rail transit equipment and storage medium
CN113807743A (en) Power grid dispatching automation software reliability assessment method and system
Saidi et al. A survey of software reliability growth model selection methods for improving reliability prediction accuracy

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