CN116611589B

CN116611589B - Power failure window period prediction method, system, equipment and medium for main network power transmission and transformation equipment

Info

Publication number: CN116611589B
Application number: CN202310887224.8A
Authority: CN
Inventors: 张风彬; 孙大雁; 李立新; 陶洪铸; 刘幸蔚; 杨军峰; 於益军; 王超; 宋旭日; 董时萌; 卫泽晨; 武力; 杨楠; 姚伟峰; 张加力; 李增辉; 黄宇鹏; 叶瑞丽; 齐晓琳; 韩昳
Original assignee: China Electric Power Research Institute Co Ltd CEPRI
Current assignee: China Electric Power Research Institute Co Ltd CEPRI
Priority date: 2023-07-19
Filing date: 2023-07-19
Publication date: 2023-09-29
Anticipated expiration: 2043-07-19
Also published as: CN116611589A

Abstract

A method, a system, a device and a medium for predicting a power failure window period of a main network power transmission and transformation device belong to the technical field of power system automation, and the prediction method comprises the following steps: main characteristic factors related to power failure of main network power transmission and transformation equipment in a power grid area are obtained, and the main characteristic factors are classified into three groups of power grid operation reliability, electric power and electric quantity balance and clean energy consumption; calculating the association degree between each main characteristic factor inside each group and the equipment through a gray association analysis method so as to determine the key characteristics of the equipment; taking sample data of the equipment in the key feature space as input quantity of a support vector machine algorithm so as to solve a power failure window period prediction model; and obtaining a power failure window period prediction result of the equipment by using the power failure window period prediction model. The invention realizes the automatic identification of the key characteristics of the power failure window period of the main network power transmission and transformation equipment, and the scientific judgment of the power failure window period of the main network power transmission and transformation equipment under the key characteristic space, thereby saving time and labor cost.

Description

Power failure window period prediction method, system, equipment and medium for main network power transmission and transformation equipment

Technical Field

The invention belongs to the technical field of power system automation, and particularly relates to a method, a system, equipment and a medium for predicting a power failure window period of power transmission and transformation equipment of a main network.

Background

The purpose of the grid plant blackable window period (simply blackout window period) is to provide one or more suitable time periods for the plant during which the arrangement of the plant blackout will have minimal impact on the overall operational reliability of the grid, the power-to-power balance and clean energy consumption. The evaluation of the power grid equipment power failure window period needs to comprehensively consider the influence of the power grid equipment power failure window period on three main factors such as power grid operation reliability, electric power and electric quantity balance, clean energy consumption and the like. In the current actual engineering scene, the equipment outage window period is mainly manually estimated by related responsibilities according to working experience, and the general process is as follows: firstly, selecting a plurality of factors which are most influenced by equipment power failure according to service experience of the past year, then checking historical data curves of the factors one by one to define a threshold value, and judging that a power failure window period is a time period within the threshold value range.

The Chinese patent application with the publication number of CN111917139A discloses a method and a system for determining a blackable window period of power grid main equipment, wherein the method comprises the steps of determining the classification type of the power transmission and transformation equipment in the power grid according to the functions of the power transmission and transformation equipment; determining and collecting basic data corresponding to the classification type according to the classification type; determining a establishment principle of a blackout window period of the main network power transmission and transformation equipment according to the basic data; and outputting and displaying the blackable window period of the main network power transmission and transformation equipment. The scheme creatively uses scientific and objective means to define the general establishment principle of the power outage window period of the power transmission and transformation equipment of the main network, so that the establishment of the power outage window period has scientific basis. However, the determination of the basic data required by the main network power transmission and transformation equipment in the scheme is too dependent on the service experience of related personnel, and the process of determining the basic data required by each equipment one by one is too complicated; the principle of determining the blackable window period according to the basic data is too wide, subjective indexes are easily interfered, and the subjective indexes are not clear and focused. At present, with the expansion of power grid business, the method for determining basic data and window period needs further innovation.

The Chinese patent application with publication number of CN113590682A discloses a method, a device, electronic equipment and a storage medium for generating a power grid power failure window period, which comprises the steps of taking operational and maintenance condition data of equipment as a reference sequence, and forming a compared sequence by related factor data which has an effect on equipment power failure; calculating association coefficients between each influence factor and power failure equipment by using the reference sequence and the comparison sequence data; calculating the association degree between each influence factor and the power failure equipment by using the association coefficient; obtaining a criterion index of a power failure window period of equipment; and verifying the allowable outage time of each power grid device through power flow calculation of the power grid by using the criterion index as a constraint condition, and generating a outage window period of the device. The scheme improves the effectiveness and safety of power grid maintenance planning, improves the working efficiency of the power outage window period, and reduces the working intensity of power outage planning staff. However, in the scheme, the process of checking the allowable outage time of each power grid device through power grid power flow calculation to generate the outage window period is not clear, and the realization path efficiency of simulation through power flow calculation is low. Therefore, the method for solving the power failure window period of the equipment needs further improvement and perfection.

Disclosure of Invention

The invention aims to solve the problems in the prior art and provides a method, a system, equipment and a medium for predicting the power failure window period of main network power transmission and transformation equipment, which realize automatic identification of key characteristics of the power failure window period of the main network power transmission and transformation equipment and scientific judgment of the power failure window period of the main network power transmission and transformation equipment in a key characteristic space, thereby saving time and labor cost.

In order to achieve the above purpose, the present invention has the following technical scheme:

in a first aspect, a method for predicting a power failure window period of a power transmission and transformation device of a main network is provided, including:

main characteristic factors related to power failure of main network power transmission and transformation equipment in a power grid area are obtained, and the main characteristic factors are classified into three groups of power grid operation reliability, electric power and electric quantity balance and clean energy consumption;

calculating the association degree between each main characteristic factor inside each group and the equipment through a gray association analysis method so as to determine the key characteristics of the equipment;

taking sample data of the equipment in the key feature space as input quantity of a support vector machine algorithm so as to solve a power failure window period prediction model;

and obtaining a power failure window period prediction result of the equipment by using the power failure window period prediction model.

As a preferred solution, the method further includes the steps of acquiring, classifying and modeling the power grid characteristic data, where the steps of acquiring, classifying and modeling the power grid characteristic data include:

determining main network power transmission and transformation equipment facing to a power grid annual maintenance plan;

acquiring characteristic data of power transmission and transformation equipment of a main network, wherein the characteristic data comprises historical blackable window periods, historical power flow data and forecast data;

modeling the set of the main network power transmission and transformation equipment and the characteristic data of the main network power transmission and transformation equipment, and calculating the association degree between each main characteristic factor and equipment in each group by using a gray association analysis method by using a mathematical expression after modeling of equipment to be evaluated.

As a preferred solution, in the step of obtaining main characteristic factors related to power outage of the power transmission and transformation equipment of the main network in the power grid area and classifying the main characteristic factors into three groups of power grid operation reliability, electric power and electric quantity balance and clean energy consumption, the main characteristic factors of the power grid operation reliability include: the method comprises the steps of associating a heavy load transmission section power flow with a power transmission end power grid which has an influence on power transmission of a main network cross-region cross-province, a power transmission end power grid main transformer on-grid section power flow with a power transmission end which has an influence on the main network cross-region cross-province, a power transmission end power grid associating heavy load transmission section power flow with a power reception end which has an influence on the main network cross-region cross-province, and a power reception end power grid main transformer off-grid section power flow with an influence on the main network cross-region cross-province power reception;

The main characteristic factors of the electric power and electric quantity balance of the power grid comprise: regional power grid dispatching caliber power generation and power generation, provincial power grid dispatching caliber power generation and power generation, local power grid dispatching caliber power generation with internal power generation blocked or focused attention;

the clean energy consumption main characteristic factors of the power grid comprise: regional power grid dispatching caliber hydroelectric power generation, regional power grid dispatching caliber photovoltaic power generation, provincial power grid dispatching caliber hydroelectric power generation, provincial power grid dispatching caliber wind power generation, provincial power grid dispatching caliber photovoltaic power generation, local power grid dispatching caliber hydroelectric power generation with external power transmission blocked or focused attention, local power grid dispatching caliber wind power generation with external power transmission blocked or focused attention, local power grid dispatching caliber photovoltaic power generation with external power transmission blocked or focused attention, national alignment water dispatching power plant dispatching caliber power generation and network provincial dispatching focused attention hydropower plant dispatching caliber power generation.

As a preferable scheme, the method further comprises the steps of obtaining historical data and prediction data of main characteristic factors of each group, modeling the main characteristic factors of each group and the corresponding historical data and prediction data, and calculating by using a mathematical expression modeled by each main characteristic factor inside each group when calculating the association degree between each main characteristic factor inside each group and equipment by a gray association analysis method.

As a preferred solution, in the step of determining the key feature of the device by calculating the association degree between each main feature factor and the device inside each group by using the gray association analysis method, the solving the association degree between the device to be evaluated and the power grid operation reliability feature factor includes:

extracting historical data sequences of all power grid operation reliability characteristic factors：

By means of a averaging method, the power grid operation reliability is classified into the first categorykPersonal characteristic factorThe history data sequence of the element is processed in a dimensionless way to obtain a reference sequenceThe calculation formula is as follows:

solving a correlation coefficient sequence between each characteristic factor and the ith equipment under the classification of the running reliability of the power grid by a gray correlation analysis method, wherein a calculation formula is as follows:

wherein ,a correlation coefficient between a jth data point representing a kth characteristic factor under the power grid operation reliability classification and an ith device; />Representing enumeration of j from 1 to M solving +.>Is set to be a minimum value of (c),representing enumeration of j from 1 to M solving +.>Is the maximum value of (2);represent enumerating k from 1 to +.>Find->Is set to be a minimum value of (c),represent enumerating k from 1 to +.>Find->Is the maximum value of (2); />A j data point in a k characteristic factor historical data sequence representing the operation reliability of the power grid,/- >Representing a j-th data point in the i-th device historical trend data sequence;

and solving the association degree between each characteristic factor and the ith equipment under the power grid operation reliability classification by weighted average, wherein the calculation formula is as follows:

ranking the association degree between each characteristic factor under the power grid operation reliability classification and the ith equipment from large to small to form an association degree sequence between the equipment Ei and all characteristic factors of the power grid operation reliability:

wherein ,representing the>The association of the individual characteristic factors with the device Ei is ranked in the kth bit.

As a preferred solution, in the step of determining the key feature of the device by calculating the association degree between each main feature factor and the device in each group through the gray association analysis method, the solving the association degree between the device to be evaluated and the power grid power and electricity balance feature factor includes:

historical data sequence for extracting all power grid power and electricity balance characteristic factors：

By means of a averaging method, the power grid power and electricity balance is classifiedkThe historical data sequence of each characteristic factor is subjected to dimensionless treatment to obtain a reference sequenceThe calculation formula is as follows:

Solving a correlation coefficient sequence between each characteristic factor and the ith equipment under the power grid electric power and electric quantity balance classification by a gray correlation analysis method, wherein a calculation formula is as follows:

wherein (1)>A correlation coefficient between a jth data point representing a kth characteristic factor under the power grid power and power balance classification and an ith device;

and solving the association degree between each characteristic factor and the ith equipment under the power grid electric power and electric quantity balance classification by weighted average, wherein the calculation formula is as follows:

the association degree between each characteristic factor under the power grid power and electricity balance classification and the ith equipment is ranked from large to small, and an association degree sequence between the equipment Ei and all the characteristic factors of the power grid power and electricity balance is formed:

wherein ,representing +.f under grid Power-quantity balance Classification>The association of the individual characteristic factors with the device Ei is ranked in the kth bit.

As a preferred solution, in the step of determining key characteristics of the device by calculating the association degree between each main characteristic factor and the device inside each group by using a gray association analysis method, the step of solving the association degree between the device to be evaluated and the power grid clean energy consumption characteristic factor includes:

historical data sequence for extracting clean energy consumption characteristic factors of all power grids ：

The method comprises the following steps of classifying clean energy consumption of the power grid by means of a averaging methodkThe historical data sequence of each characteristic factor is subjected to dimensionless treatment to obtain a reference sequenceThe calculation formula is as follows:

solving a correlation coefficient sequence between each characteristic factor and the ith equipment under the power grid clean energy consumption classification by a gray correlation analysis method, wherein a calculation formula is as follows:

wherein ,a correlation coefficient between a jth data point representing a kth characteristic factor under the power grid clean energy consumption classification and an ith device;

and solving the association degree between each characteristic factor and the ith equipment under the power grid clean energy consumption classification by weighted average, wherein the calculation formula is as follows:

the association degree between each characteristic factor under the power grid clean energy consumption classification and the ith equipment is ranked from large to small, and an association degree sequence between the equipment Ei and all the characteristic factors of the power grid clean energy consumption is formed:

wherein ,represents +.f under the grid clean energy consumption category>The association of the individual characteristic factors with the device Ei is ranked in the kth bit.

As a preferred solution, in the step of determining the key features of the device by calculating the degree of association between each main feature factor inside each group and the device by the gray correlation analysis method, determining the key features of the device to be evaluated includes:

The running reliability of the power grid, the electric power and electric quantity balance, and the overall association degree of clean energy consumption and equipment Ei to be evaluated are respectively as follows:

defining the space dimension of key characteristics of equipment Ei to be evaluated as w dimension according to the following steps of、/>、/>The weight ratio of the equipment to be evaluated Ei is calculated, and the operation reliability of the power grid, the electric power and electric quantity balance and the clean energy consumption are respectively as follows:

wherein int represents a downward rounding function;

the key feature space of each group of equipment Ei to be evaluated for determining the operation reliability of the power grid, the electric power and electric quantity balance and clean energy consumption is characterized in that:

wherein ,a key feature space representing the device to be evaluated Ei, < ->Representing the>Individual characteristic factors->Representing +.f under grid Power-quantity balance Classification>Individual characteristic factors->Represents +.f under the grid clean energy consumption category>And a characteristic factor.

As a preferred solution, the step of solving the outage window prediction model by using the sample data of the device in the key feature space as the input quantity of the support vector machine algorithm includes:

key feature space of equipment Ei to be evaluated The history data sequence of all characteristic factors in (a) as a characteristic data column +.>Taking the historical blackout window period data sequence of the equipment Ei to be evaluated as a marking data column +.>The sample data standard input set of the equipment Ei suitable for the SVM model is constructed as follows:

dividing the sampled data points in the device Ei sample data standard input set into a training set +.>And test set->；

And (3) taking a Gaussian function as a kernel function of an SVM algorithm, and expressing a blackout window period SVM prediction model of the equipment Ei to be evaluated as:

wherein ,is Gaussian kernel parameter->For model loss factor, +.>Outputting a result of the power failure window period prediction model of the equipment Ei;

setting an initial stage of SVM trainingIteration parameter->Iterative step size->The method comprises the steps of carrying out a first treatment on the surface of the Setting an initial valueIteration parameter->Iterative step size->；

The accuracy of the training result obtained by training the SVM model under various parameter combinations is as follows:

wherein ,the accuracy of the result of the k-th SVM model training of the equipment Ei is represented, and L1 represents the total number of initial model training of the round;

selecting Gaussian kernel parameters with highest accuracy in initial model training resultsAnd model loss factor->Combinations, respectively defined as->、/>；

To be used for and />Setting optimization +.>Iteration parameter->Iterative step size- >The method comprises the steps of carrying out a first treatment on the surface of the Setting optimization->Iteration parameter->Iterative step size->The method comprises the steps of carrying out a first treatment on the surface of the Will optimize->And optimize->Is used as the input parameter of the model, and is trained again to obtain the accuracy of the resultThe ratio is:

wherein, L2 represents the total number of times of the round of optimization model training;

the Gaussian kernel parameter and the model loss coefficient with the highest accuracy in the training result of the L2 round optimization model are respectively defined as、/>The optimal prediction model of the power failure window period of the equipment Ei to be evaluated is obtained by the method:

。

as a preferred embodiment, the step of obtaining the outage window prediction result of the device using the outage window prediction model includes:

key feature space of equipment Ei to be evaluatedThe predicted data sequence of all characteristic factors in the equipment Ei is used as a characteristic data sequence, and a characteristic data set of the equipment Ei is constructed as follows:

wherein P is the total number of data points to be evaluated;

feature data set of equipment Ei to be evaluatedAs->Is solved to obtain the device Ei in the predicted input data matrix +.>The following predicted result output sequences are:

wherein ,the output of the result corresponding to the j-th moment predicted data point of the equipment Ei is shown, 1 shows that the current moment can be used as a power outage window, and 0 shows that the current moment cannot be used as the power outage window; combining all continuous time points which can be used as power outage windows to form a power outage window period of the equipment Ei, and representing the power outage window period as:

wherein ,and a time period is represented, t represents a certain time point in the time period, the prediction result of the corresponding data point of the time point is 1, and t represents the total number of segments of the power failure window period of the equipment Ei.

In a second aspect, a power outage window prediction system for a power transmission and transformation device of a main network is provided, including:

the main characteristic factor acquisition module is used for acquiring main characteristic factors related to power failure of the main network power transmission and transformation equipment in the power grid area and classifying the main characteristic factors into three groups of power grid operation reliability, electric power and electric quantity balance and clean energy consumption;

the key feature determining module is used for calculating the association degree between each main feature factor inside each group and the equipment through a gray association analysis method so as to determine key features of the equipment;

the power failure window period prediction model solving module is used for taking sample data of the equipment in the key feature space as input quantity of a support vector machine algorithm so as to solve a power failure window period prediction model;

and the prediction module is used for obtaining a power failure window period prediction result of the equipment by using the power failure window period prediction model.

In a third aspect, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements the method for predicting a power outage window period of a power transmission and transformation device of a main network when executing the computer program.

In a fourth aspect, a computer readable storage medium is provided, where the computer readable storage medium stores a computer program, where the computer program when executed by a processor implements the method for predicting a power outage window period of a power transmission and transformation device of a main network.

Compared with the prior art, the first aspect of the invention has at least the following beneficial effects:

main characteristic factors related to power failure of main network power transmission and transformation equipment in a power grid area are classified into three groups of power grid operation reliability, electric power and electricity balance and clean energy consumption, so that the problems of over conceptualization of the establishment basis of the existing power failure window period and insufficient definition and focusing are avoided. According to the calculation flow of the key features of the equipment to be evaluated, the overall association degree of the equipment to be evaluated with the operation reliability of the power grid, the electric power and electric quantity balance and the clean energy consumption is accurately calculated through mathematical modeling, the quick solution of the key feature space of the equipment is realized, and the problems that the judgment of the key features of the power failure window period of the current power grid equipment is excessively dependent on service experience, is slow and tedious are solved. The invention combines a gray correlation analysis (Grey Relation Analysis, GRA) method and a support vector machine (Support Vector Machines, SVM) classification model, GRA is a multi-factor correlation analysis method, which can clearly calculate the relative strength of a concerned object affected by other factors, thereby finding the most relevant influence factor of the object. The SVM is a feature space-based two-class model, which can perform model training according to input sample feature data, and can accurately separate subsequently input samples to be evaluated by utilizing training results. The invention provides a method for solving a blackout window period prediction model under a key feature space of power grid equipment, and the provided coarse-before-fine dual-mode iterative method searches for optimized model parameters, so that a blackout window period optimization prediction model of the equipment can be better obtained, and the blackout window period of the equipment can be rapidly judged.

It will be appreciated that the advantages of the second to fourth aspects may be found in the relevant description of the first aspect and are not repeated here.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flow chart of a method for predicting a power failure window period of a main network power transmission and transformation device in an embodiment of the application;

fig. 2 is a block diagram of a power failure window period prediction system of a main network power transmission and transformation device according to an embodiment of the present application.

Detailed Description

In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system configurations, techniques, etc. in order to provide a thorough understanding of embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.

Furthermore, the terms "first," "second," "third," and the like in the description of the present specification and in the appended claims, are used for distinguishing between descriptions and not necessarily for indicating or implying a relative importance.

Referring to fig. 1, the method for predicting the outage window period of the power transmission and transformation equipment of the main network according to the embodiment of the application realizes the automatic identification of key features of the outage window period of the power transmission and transformation equipment of the main network by using a gray correlation analysis method improved by grouping and aggregation, the method and the system for determining the blackable window period of the power grid equipment can solve the problems that key characteristics or basic data of the blackable window period of the power grid equipment in the patent application (publication number is CN 111917139A) depend on service experience excessively and the process is complicated excessively; secondly, through a support vector machine algorithm with excellent classification performance, scientific judgment of the blackout window period of the main network power transmission and transformation equipment under the key feature space is realized, and the problems that the principle of making the blackout window period in the power grid blackout window period generating method, device, electronic equipment and storage medium (with the publication number of CN 113590682A) is excessively conceptual, and focusing is insufficient or a basis is not scientifically made in the patent application are solved. Secondly, the problems with current engineering practices are: firstly, according to accumulated service experience, related personnel combine main factors related to power failure of power grid equipment into three types of power grid operation reliability, electric power and electric quantity balance and clean energy consumption, but no related technical solution exists for the factors contained in each type at present; secondly, the most relevant key features of the equipment under the three groups of classification still depend on service experience excessively in actual service at present, and the prior art has a detailed solution to the problem (a patent application of a power grid outage window period generation method, a device, an electronic device and a storage medium (the publication number is mentioned in CN 113590682A); thirdly, judging the blackout window period of the equipment under the key feature space, wherein the actual business is currently defined by transiting experience according to the data curve of the key feature one by one, the manual process has extremely time and labor consumption, and the prior art has no scientific solution to the problem.

The method for predicting the power failure window period of the power transmission and transformation equipment of the main network mainly comprises the following steps:

s1, acquiring main characteristic factors related to power failure of main network power transmission and transformation equipment in a power grid area, and classifying the main characteristic factors into three groups of power grid operation reliability, electric power and electric quantity balance and clean energy consumption;

s2, calculating the association degree between each main characteristic factor inside each group and the equipment through a gray association analysis method so as to determine the key characteristics of the equipment;

s3, taking sample data of the equipment in the key feature space as input quantity of a support vector machine algorithm so as to solve a power failure window period prediction model;

s4, obtaining a power failure window period prediction result of the equipment by using the power failure window period prediction model.

In one possible implementation manner, the method of this embodiment further includes the steps of acquiring, classifying and modeling the power grid characteristic data, and further, the steps of acquiring, classifying and modeling the power grid characteristic data include:

(1) Determining main network power transmission and transformation equipment facing to a power grid annual maintenance plan;

mainly comprises the following steps: the system comprises a main network cross-region alternating current circuit, a main network cross-region direct current system, a main network 220kV and above voltage class intra-region alternating current circuit, a main network 220kV and above voltage class intra-region direct current system, a main network 220kV and above voltage class bus, a main network 220kV and above voltage class transformer, a main network 220kV and above voltage class converter, a main network cross-region direct current grounding system, other water, fire core wind and light split type power supply outgoing channel equipment, other equipment which has influence on the main network cross-region power transmission capability and the like.

(2) The method comprises the steps of obtaining characteristic data of power transmission and transformation equipment of a main network, wherein the characteristic data comprise historical blackable window periods, historical power flow data and prediction data, and specifically comprises the following steps:

the method for acquiring the power transmission and transformation equipment of the main network and the historical blackable window period thereof mainly comprises the following steps:

1) A main network cross-region alternating current circuit and a historical blackable window period thereof;

2) A main network cross-region direct current system and a historical blackable window period thereof;

3) The main network spans the provincial alternating current line and the historical blackable window period thereof;

4) The main network spans the provincial direct current system and the historical blackable window period thereof;

5) An intra-provincial alternating current circuit with a voltage class of 220kV and above of the main network and a historical blackable window period thereof;

6) An intra-provincial direct current system with a voltage class of 220kV and above of the main network and a historical blackable window period thereof;

7) Bus with 220kV and above voltage class of main network and its historical power failure window period;

8) A transformer with a voltage class of 220kV and above of the main network and a historical blackable window period thereof;

9) A converter with a voltage class of 220kV and above of the main network and a historical blackable window period thereof;

10 A main network cross-region and cross-province direct current grounding system and a historical power failure window period of the direct current system to which the main network belongs;

11 Other water and fire nuclear wind and light split type power supply output channel equipment and historical blackable window periods thereof;

12 Other devices that have an impact on the main network's cross-zone cross-power save capability and its historical blackout window period.

The time interval of data sampling is 1 hour, and the time length of data sampling is the last three complete years. And if the sampling point is defined as 1 during the blackout window period of the device, if the sampling point is not defined as 0 during the blackout window period of the device.

The method for acquiring the power transmission and transformation equipment of the main network and the historical power flow data thereof mainly comprises the following steps:

1) Active power is sampled by a main network cross-region alternating current circuit and a transmitting end thereof;

2) The main network cross-region direct current system and the sending end thereof sample active power;

3) The main network cross-provincial alternating current circuit and the sending end thereof sample active power;

4) The main network cross-provincial direct current system and the sending end thereof sample active power;

5) An intra-provincial alternating current circuit with the voltage class of 220kV and above of the main network and a transmitting end thereof sample active power;

6) The provincial direct current system with the voltage class of 220kV and above of the main network and the sending end thereof sample active power;

7) Bus bars with voltage class of 220kV and above of the main network and active power thereof;

8) A transformer with a voltage class of 220kV and above of a main network and active power sampled at a high-voltage side of the transformer;

9) Converter with voltage class of 220kV and above of main network and active power thereof;

10 Main network cross-region cross-province direct current grounding system and the sending end of the direct current system to which the main network belongs sample active power;

11 Other water and fire core wind and light division type power supply output channel equipment and its output end sampling active power;

12 Other devices that have an impact on the main network's trans-regional trans-provincial power delivery capability and its historical active power.

The statistical caliber of the data sampling is a scheduling caliber or other uniform calibers, the time interval of the data sampling is 1 hour, and the time length of the data sampling is the last three complete years.

The method for acquiring the power transmission and transformation equipment of the main network and the prediction data thereof mainly comprises the following steps:

1) A main network cross-region alternating current circuit and a sending end active power predicted value thereof;

2) Main network cross-region direct current system and its transmitting end active power predicted value;

3) The main network spans the provincial alternating current line and the active power predicted value of the sending end thereof;

4) The main network spans the provincial direct current system and the active power predicted value of the sending end thereof;

5) An intra-provincial alternating current circuit with the voltage class of 220kV and above of the main network and a sending end active power predicted value thereof;

6) An intra-provincial direct current system with the voltage class of 220kV and above of the main network and a sending end active power predicted value thereof;

7) Bus bars with voltage class of 220kV and above of the main network and active power predicted values thereof;

8) A transformer with a voltage class of 220kV and above of the main network and a high-voltage side active power predicted value thereof;

9) A converter with a voltage class of 220kV and above of the main network and an active power predicted value thereof;

10 Main network cross-region and cross-province direct current grounding system and a transmitting end active power predicted value of the direct current system to which the main network belongs;

11 Other water and fire core wind and light split type power supply output channel equipment and its output active power predictive value;

12 Other devices that have an impact on the main network's trans-regional trans-proving power delivery capability and active power predictions.

The statistical caliber of the data prediction is a scheduling caliber or other uniform calibers, the time interval of the data prediction is 1 hour, and the time length of the data prediction is the time range for carrying out the power failure window period assessment.

(3) Modeling the set of the main network power transmission and transformation equipment and the characteristic data of the main network power transmission and transformation equipment, and calculating the association degree between each main characteristic factor and equipment in each group by using a gray association analysis method by using a mathematical expression after modeling of equipment to be evaluated.

Defining a set of main network power transmission and transformation equipment as follows:

wherein ,representing an ith device; n represents the total number of the power transmission and transformation equipment of the main network.

Defining historical blackout window period data of the main network power transmission and transformation equipment as:

wherein N is the total number of the power transmission and transformation equipment of the main network,a historical blackout window period data sequence representing an i-th device, defined as:

wherein ,and representing the j-th data point in the data sequence of the i-th equipment history power failure window period, wherein the data points in the data sequence are orderly arranged from front to back according to the sampling time, M represents the total number of the historical data sampling points, and N is the total number of the main network power transmission and transformation equipment.

Defining the historical power flow data of the main network power transmission and transformation equipment as:

wherein N is the total number of the power transmission and transformation equipment of the main network,a historical power flow data sequence representing an i-th device, defined as:

wherein ,and representing the jth data point in the historical power flow data sequence of the ith device, wherein the data points in the data sequence are orderly arranged from front to back according to sampling time, M is the total number of the historical data sampling points, and N is the total number of the power transmission and transformation devices of the main network.

The prediction data of the power transmission and transformation equipment of the main network are defined as:

wherein N is the total number of the power transmission and transformation equipment of the main network,a predicted data sequence representing an i-th device, defined as:

wherein ,the ith device predicts the jth data point in the data sequence, and the data points in the data sequence should be arranged in sequence from front to back according to the sampling time, P represents the total number of the predicted data sampling points, and N is the total number of the power transmission and transformation devices of the main network.

In one possible implementation, step S1 classifies the main characteristic factors into three groups of grid operation reliability, electric power and electricity balance and clean energy consumption, where the main characteristic factors of grid operation reliability include: the method comprises the steps of associating a heavy load transmission section power flow with a power transmission end power grid which has an influence on power transmission of a main network cross-region cross-province, a power transmission end power grid main transformer on-grid section power flow with a power transmission end which has an influence on the main network cross-region cross-province, a power transmission end power grid associating heavy load transmission section power flow with a power reception end which has an influence on the main network cross-region cross-province, and a power reception end power grid main transformer off-grid section power flow with an influence on the main network cross-region cross-province power reception;

In one possible implementation manner, the method of the present embodiment further includes obtaining historical data and prediction data of each group of main feature factors, modeling each group of main feature factors and corresponding historical data and prediction data, and calculating, when calculating the degree of association between each main feature factor inside each group and the device by using the gray association analysis method, using the mathematical expression after modeling each main feature factor inside each group.

The method for acquiring the historical data of the main characteristic factors of the operation reliability of the power grid mainly comprises the following steps:

the channel equipment of the power transmission end power grid, which has influence on the cross-region and cross-power-saving transmission of the main network, and the power transmission active power of the power transmission end power grid mainly consider channel equipment such as the associated heavy-load transmission section of the power transmission end power grid, the main transformer network section of the power transmission end power grid and the like;

the channel equipment of the receiving end power grid and the power receiving active power thereof, which have influence on the cross-regional and cross-proving power receiving of the main network, mainly consider channel equipment such as the associated heavy-load transmission section of the receiving end power grid, the main transformer off-grid section of the receiving end power grid and the like.

The method for acquiring the prediction data of the main characteristic factors of the operation reliability of the power grid mainly comprises the following steps:

the channel equipment of the power transmission end power grid, which has influence on the cross-region, cross-saving and power transmission of the main network, and the power transmission active power prediction value of the power transmission end power grid mainly consider channel equipment such as a heavy load transmission section, a main transformer network section of the power transmission end power grid and the like;

the channel equipment of the receiving end power grid, which has influence on the power supply of the main network in a cross-region and cross-province manner, and the power supply active power prediction value of the receiving end power grid mainly consider channel equipment such as a heavy load power transmission section, a main transformer power supply section of the receiving end power grid and the like.

The method for acquiring the historical data of the main characteristic factors of the electric power and electric quantity balance of the power grid mainly comprises the following steps:

regional power grid dispatching caliber power generation and generation power;

the provincial power grid schedules caliber power generation and power generation;

the local power grid dispatching caliber generated power which is blocked in internal power reception or focused on is stored.

The method for acquiring the prediction data of the main characteristic factors of the electric power and electric quantity balance of the power grid mainly comprises the following steps:

regional power grid dispatching caliber power generation and generation power prediction values;

provincial power grid dispatching caliber power generation and generation power prediction value;

the prediction value of the power generation and generation power of the dispatching caliber of the local power grid, which is blocked or focused on by internal power reception, exists.

The method for acquiring the historical data of the main characteristic factors of the clean energy consumption of the power grid mainly comprises the following steps:

regional power grid dispatching caliber hydroelectric power generation output;

regional power grid dispatching caliber wind power generation output;

regional power grid dispatching caliber photovoltaic power generation output;

the provincial power grid dispatches caliber hydroelectric power generation output;

the provincial power grid dispatches caliber wind power generation output;

the provincial power grid schedules caliber photovoltaic power generation output;

the caliber hydroelectric power generation output of the dispatching caliber of the local power grid with blocked external power transmission or focused attention exists;

the wind power generation output of the caliber is scheduled by a local power grid with blocked external power transmission or focused attention;

the photovoltaic power generation capacity of the dispatching caliber of the local power grid with blocked external power transmission or focused attention exists;

Dispatching caliber power generation output of the national straightening water-regulating power plant;

the power generation capacity of the dispatching caliber of the hydropower plant is focused on by the network province dispatching center.

The method for acquiring the forecast data of the main characteristic factors of the clean energy consumption of the power grid mainly comprises the following steps:

regional power grid dispatching caliber hydroelectric power generation output predicted value;

regional power grid dispatching caliber wind power generation output predicted value;

regional power grid dispatching caliber photovoltaic power generation output predicted value;

provincial power grid dispatching caliber hydroelectric power generation output predicted value;

dispatching caliber wind power generation output predicted value of provincial power grid;

provincial power grid dispatching caliber photovoltaic power generation output predicted value;

the predicted value of the hydroelectric power generation output of the dispatching caliber of the local power grid with blocked external power transmission or focused attention exists;

the wind power generation output predicted value of the dispatching caliber of the local power grid with blocked external power transmission or focused attention exists;

the predicted value of the photovoltaic power generation output of the dispatching caliber of the local power grid with blocked external power transmission or focused attention exists;

a predicted value of the power generation output of the dispatching caliber of the national straightening water-regulating power plant;

The power generation output predicted value of the dispatching caliber of the hydropower plant is focused on by the network province dispatching.

Modeling main characteristic factors and data of a power grid in the following manner;

the main characteristic factors of the operation reliability of the power grid are defined as follows:

wherein ,representing the kth characteristic factor under the power grid operation reliability classification; />And the total number of the characteristic factors of the running reliability of the power grid is represented.

The historical data of the main characteristic factors of the operation reliability of the power grid are defined as follows:

wherein ,for the total number of the above-mentioned power grid operational reliability characteristic factors, +.>A historical data sequence representing a kth characteristic factor of the operation reliability of the power grid is defined as:

wherein ,the method comprises the steps that the jth data point in a kth characteristic factor historical data sequence of the operation reliability of the power grid is represented, and the data points in the data sequence are sequentially arranged from front to back according to sampling time; m represents the total number of historical data sampling points, < ->The total number of the characteristic factors of the operation reliability of the power grid is obtained.

The prediction data of the main characteristic factors of the operation reliability of the power grid are defined as follows:

wherein ,for the total number of the above-mentioned power grid operational reliability characteristic factors, +.>A predicted data sequence representing a kth characteristic factor of the operation reliability of the power grid is defined as: />

wherein ,the method comprises the steps that the kth characteristic factor representing the running reliability of a power grid predicts the jth data point in a data sequence, and the data points in the data sequence are orderly arranged from front to back according to sampling time; m represents the total number of predicted data samples, < >>The total number of the characteristic factors of the operation reliability of the power grid is obtained.

The main characteristic factors of the electric power and electric quantity balance of the power grid are defined as follows:

wherein ,representing the kth characteristic factor under the power grid power and electricity balance classification; />Representing an electrical gridTotal number of power and energy balance characteristic factors.

The historical data of main characteristic factors of the electric power and electric quantity balance of the power grid are defined as:

wherein ,the total number of the power grid power and electric quantity balance characteristic factors is +.>A historical data sequence representing the kth characteristic factor of the power and electric quantity balance of the power grid is defined as:

wherein ,the method comprises the steps that the jth data point in a kth characteristic factor historical data sequence for representing electric power and electric quantity balance of a power grid is represented, and the data points in the data sequence are sequentially arranged from front to back according to sampling time; m represents the total number of historical data sampling points, < - >The total number of the characteristic factors of the electric power and electric quantity balance of the power grid is obtained.

The prediction data of main characteristic factors of the electric power and electric quantity balance of the power grid are defined as follows:

wherein ,the total number of the power grid power and electric quantity balance characteristic factors is +.>A predicted data sequence representing a kth characteristic factor of power and electric quantity balance of the power grid is defined as:

wherein ,the method comprises the steps that a kth characteristic factor for representing electric power and electric quantity balance of a power grid predicts a jth data point in a data sequence, and the data points in the data sequence are sequentially arranged from front to back according to sampling time; m represents the total number of predicted data samples, < >>The total number of the characteristic factors of the electric power and electric quantity balance of the power grid is obtained.

The main characteristic factors of clean energy consumption of the power grid are defined as follows:

wherein ,representing the kth characteristic factor under the clean energy consumption classification of the power grid; />And the total number of the clean energy consumption characteristic factors of the power grid is represented.

The historical data of the main characteristic factors of the clean energy consumption of the power grid are defined as follows:

wherein ,clean energy consumption features for the above-mentioned power gridTotal number of factors>A historical data sequence representing the k-th characteristic factor of clean energy consumption of the power grid is defined as:

wherein ,representing the jth data point in the kth characteristic factor historical data sequence of the clean energy consumption of the power grid, and arranging the data points in the data sequence from front to back according to the sampling time; m represents the total number of historical data sampling points, < - >And (5) the total number of characteristic factors for the clean energy consumption of the power grid. />

The prediction data of main characteristic factors of clean energy consumption of the power grid are defined as follows:

wherein ,for the total number of the characteristic factors of the clean energy consumption of the power grid, < >>A predicted data sequence representing the k-th characteristic factor of clean energy consumption of the power grid is defined as:

wherein ,represents the k-th clean energy consumption of the power gridPredicting the jth data point in the data sequence by the characteristic factors, and arranging the data points in the data sequence from front to back according to the sampling time; m represents the total number of predicted data samples, < >>And (5) the total number of characteristic factors for the clean energy consumption of the power grid.

In one possible embodiment, in step S2, an example of solving the key features of the ith main network power transmission and transformation device to be evaluated is described.

Solving a dimensionless data sequence of the device under evaluation as follows:

extracting historical tide data sequence of ith equipmentThe method comprises the following steps:

the historical trend data sequence of the ith equipment is subjected to dimensionless treatment by a mean value method to obtain a reference sequenceThe calculation formula is as follows:

the method for solving the association degree of the equipment to be evaluated and the power grid operation reliability characteristic factors comprises the following steps:

By means of the averaging method,classifying the operation reliability of the power gridkThe historical data sequence of each characteristic factor is subjected to dimensionless treatment to obtain a reference sequenceThe calculation formula is as follows:

wherein ,/>A correlation coefficient between a jth data point representing a kth characteristic factor under the power grid operation reliability classification and an ith device;representing enumeration of j from 1 to M solving +.>Minimum value->Representing enumeration of j from 1 to M solving +.>Is the maximum value of (2); />Represent enumerating k from 1 to +.>ObtainingMinimum value->Represent enumerating k from 1 to +.>Find->Is the maximum value of (2); />A j data point in a k characteristic factor historical data sequence representing the operation reliability of the power grid,/->Representing a j-th data point in the i-th device historical trend data sequence;

The method for solving the association degree of the equipment to be evaluated and the power grid electric power and electric quantity balance characteristic factors comprises the following steps:

wherein ,/>A correlation coefficient between a jth data point representing a kth characteristic factor under the power grid power and power balance classification and an ith device;

The method for solving the association degree of the equipment to be evaluated and the clean energy consumption characteristic factors of the power grid comprises the following steps:

historical data sequence for extracting clean energy consumption characteristic factors of all power grids：/>

Determining key features of the device under evaluation includes:

defining the space dimension of key characteristics of equipment Ei to be evaluated as w dimension according to the following steps of、/>、/>The weight ratio of the equipment to be evaluated Ei is calculated, and the operation reliability of the power grid, the electric power and electric quantity balance and the clean energy consumption are respectively as follows: />

Wherein int represents a downward rounding function;

In one possible implementation manner, step S3 is described by taking solving an i-th power outage window prediction model of the main network power transmission and transformation device to be evaluated as an example, where step S3 mainly includes:

Key feature space of equipment Ei to be evaluatedThe history data sequence of all characteristic factors in (a) as a characteristic data column +.>Taking the historical blackout window period data sequence of the equipment Ei to be evaluated as a marking data column +.>The sample data standard input set of the equipment Ei suitable for the SVM model is constructed as follows:

wherein ,is Gaussian kernel parameter->Loss for modelCoefficient of->Outputting a result of the power failure window period prediction model of the equipment Ei;

setting an initial stage of SVM trainingIteration parameter->Iterative step size->The method comprises the steps of carrying out a first treatment on the surface of the Setting the initial->Iteration parameter->Iterative step size->；

To be used for and />Setting optimization +. >Iteration parameter->Iterative step size->The method comprises the steps of carrying out a first treatment on the surface of the Setting optimization->Iteration parameter->Iterative step size->The method comprises the steps of carrying out a first treatment on the surface of the Will optimize->And optimize->The parameter combination of the model is used as the input parameter of the model, and the accuracy of the obtained result is as follows:

。

in a possible implementation manner, step S4 is illustrated by taking a solution of a power outage window prediction result of the ith main network power transmission and transformation device to be evaluated as an example, and specifically includes:

wherein P is the total number of data points to be evaluated;

And repeating the steps S2 to S4 to obtain the power failure window period prediction results of all the devices.

The invention also provides a power failure window period prediction system of the main network power transmission and transformation equipment, which comprises the following steps:

The invention also provides an electronic device, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the power failure window period prediction method of the main network power transmission and transformation device is realized when the processor executes the computer program.

Another embodiment of the present invention further provides a computer readable storage medium, where a computer program is stored, where the computer program when executed by a processor implements the method for predicting a power outage window period of a power transmission and transformation device of a main network.

The computer program comprises computer program code which may be in source code form, object code form, executable file or in some intermediate form, etc. The computer readable storage medium may include: any entity or device, medium, usb disk, removable hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier wave signals, telecommunications signals, software distribution media, and the like capable of carrying the computer program code. It should be noted that the computer readable medium contains content that can be appropriately scaled according to the requirements of jurisdictions in which such content is subject to legislation and patent practice, such as in certain jurisdictions in which such content is subject to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals. For convenience of description, the foregoing disclosure shows only those parts relevant to the embodiments of the present invention, and specific technical details are not disclosed, but reference is made to the method parts of the embodiments of the present invention. The computer readable storage medium is non-transitory and can be stored in a storage device formed by various electronic devices, and can implement the execution procedure described in the method according to the embodiment of the present invention.

It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention 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 invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flowchart and/or block of the flowchart illustrations and/or block diagrams, and combinations of flowcharts and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

Finally, it should be noted that: the above embodiments are only for illustrating the technical aspects of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the above embodiments, it should be understood by those of ordinary skill in the art that: modifications and equivalents may be made to the specific embodiments of the invention without departing from the spirit and scope of the invention, which is intended to be covered by the claims.

Claims

1. A power failure window period prediction method for a main network power transmission and transformation device is characterized by comprising the following steps:

obtaining a power failure window period prediction result of the equipment by using a power failure window period prediction model;

the method also comprises the steps of acquiring, classifying and modeling the power grid characteristic data, wherein the steps of acquiring, classifying and modeling the power grid characteristic data comprise the following steps:

modeling a set of main network power transmission and transformation equipment and characteristic data of the main network power transmission and transformation equipment, and calculating the association degree between each main characteristic factor inside each group and equipment by using a mathematical expression after modeling of equipment to be evaluated when calculating the association degree between each main characteristic factor and equipment by using a gray association analysis method;

The method comprises the steps of obtaining historical data and prediction data of main characteristic factors of each group, modeling the main characteristic factors of each group and corresponding historical data and prediction data, and calculating by using mathematical expressions after modeling of each main characteristic factor of each group when calculating the association degree between each main characteristic factor of each group and equipment by a gray association analysis method;

in the step of determining key features of the device by calculating the degree of association between each main feature factor inside each group and the device by the gray association analysis method, the determining key features of the device to be evaluated includes:

wherein int represents a downward rounding function;

wherein ,a key feature space representing the device to be evaluated Ei, < ->Representing the first power grid operational reliability categoryIndividual characteristic factors->Representing +.f under grid Power-quantity balance Classification>A number of characteristic factors are included, such as,represents +.f under the grid clean energy consumption category>A characteristic factor;

the step of obtaining the power outage window prediction result of the equipment by using the power outage window prediction model comprises the following steps:

wherein P is the total number of data points to be evaluated;

the equipment Ei to be evaluatedFeature data setAs->Is solved to obtain the device Ei in the predicted input data matrix +.>The following predicted result output sequences are:

2. The method for predicting a power outage window period of a power transmission and transformation device of a main network according to claim 1, wherein in the step of obtaining main characteristic factors related to power outage of the power transmission and transformation device of the main network in a power grid area and classifying the main characteristic factors into three groups, the power grid operation reliability, the power and electricity balance and the clean energy consumption, the main characteristic factors of the power grid operation reliability include: the method comprises the steps of associating a heavy load transmission section power flow with a power transmission end power grid which has an influence on power transmission of a main network cross-region cross-province, a power transmission end power grid main transformer on-grid section power flow with a power transmission end which has an influence on the main network cross-region cross-province, a power transmission end power grid associating heavy load transmission section power flow with a power reception end which has an influence on the main network cross-region cross-province, and a power reception end power grid main transformer off-grid section power flow with an influence on the main network cross-region cross-province power reception;

3. The method for predicting a blackout window period of a main network power transmission and transformation device according to claim 1, wherein in the step of determining key characteristics of the device by calculating the association degree between each main characteristic factor and the device inside each group through a gray association analysis method, the step of solving the association degree between the device to be evaluated and the power grid operation reliability characteristic factor comprises the following steps:

By means of a averaging method, the power grid operation reliability is classified into the first categorykThe historical data sequence of each characteristic factor is subjected to dimensionless treatment to obtain a reference sequenceThe calculation formula is as follows:

wherein ,a correlation coefficient between a jth data point representing a kth characteristic factor under the power grid operation reliability classification and an ith device; />Representing enumeration of j from 1 to M solving +.>Is set to be a minimum value of (c),representing enumeration of j from 1 to M solving +.>Is the maximum value of (2); />Representing the presentation to bek is enumerated from 1 to->Find->Minimum value->Represent enumerating k from 1 to Find->Is the maximum value of (2); />A j data point in a k characteristic factor historical data sequence representing the operation reliability of the power grid,/->Representing a j-th data point in the i-th device historical trend data sequence;

4. The method for predicting a blackout window period of a main power transmission and transformation device according to claim 1, wherein in the step of determining key characteristics of the device by calculating the association degree between each main characteristic factor and the device inside each group through a gray association analysis method, the step of solving the association degree between the device to be evaluated and the power grid power and electricity balance characteristic factor comprises the following steps:

historical data sequence for extracting all power grid power and electricity balance characteristic factors ：

wherein ,a correlation coefficient between a jth data point representing a kth characteristic factor under the power grid power and power balance classification and an ith device;

5. The method for predicting a blackout window period of a main power transmission and transformation device according to claim 1, wherein in the step of determining key characteristics of the device by calculating the association degree between each main characteristic factor and the device inside each group through a gray association analysis method, the step of solving the association degree between the device to be evaluated and the clean energy consumption characteristic factor of the power grid comprises the following steps:

Extracting clean energy absorption power of all power gridsHistorical data sequence of symptom factors：

6. The method for predicting the outage window period of the power transmission and transformation equipment of the main network according to claim 1, wherein the step of solving the outage window period prediction model by taking sample data of the equipment in a key feature space as an input quantity of a support vector machine algorithm comprises the following steps:

Key feature space of equipment Ei to be evaluatedHistorical data sequence of all characteristic factors in the database as characteristic data sequenceTaking the historical blackout window period data sequence of the equipment Ei to be evaluated as a marking data column +.>The sample data standard input set of the equipment Ei suitable for the SVM model is constructed as follows:

dividing sampled data points in a device Ei sample data standard input set into trainingCollection setAnd test set->；

setting an initial stage of SVM trainingIteration parameter->Iterative step size->The method comprises the steps of carrying out a first treatment on the surface of the Setting the initial->Iteration parametersIterative step size->；

To be used for and />Setting optimization +. >Iteration parameter->Iterative step size->The method comprises the steps of carrying out a first treatment on the surface of the Setup optimizationIteration parameter->Iterative step size->The method comprises the steps of carrying out a first treatment on the surface of the Will optimize->And optimize->The parameter combination of the model is used as the input parameter of the model, and the accuracy of the obtained result is as follows:

。

7. a power outage window prediction system for a power transmission and transformation device of a main network, which is used for realizing the power outage window prediction method for the power transmission and transformation device of the main network according to any one of claims 1-6, comprising:

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized by: the method for predicting the power failure window period of the power transmission and transformation equipment of the main network according to any one of claims 1 to 6 is realized when the processor executes the computer program.

9. A computer-readable storage medium storing a computer program, characterized in that: the computer program when executed by a processor implements a method for predicting a power outage window period of a power transmission and transformation device of a main network according to any one of claims 1 to 6.