CN113204926A - Electric appliance prediction method, device and equipment based on scoring mechanism - Google Patents

Electric appliance prediction method, device and equipment based on scoring mechanism Download PDF

Info

Publication number
CN113204926A
CN113204926A CN202110631295.2A CN202110631295A CN113204926A CN 113204926 A CN113204926 A CN 113204926A CN 202110631295 A CN202110631295 A CN 202110631295A CN 113204926 A CN113204926 A CN 113204926A
Authority
CN
China
Prior art keywords
electrical
classification model
tested
target
equipment
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
CN202110631295.2A
Other languages
Chinese (zh)
Inventor
俞啸玲
郭强
王辉东
姚海燕
胡翔
留毅
韩辉
缪忠杰
屈雷涛
孔亚广
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.)
Hangzhou Dianzi University
Hangzhou Power Equipment Manufacturing Co Ltd
Hangzhou Power Supply Co of State Grid Zhejiang Electric Power Co Ltd
Original Assignee
Hangzhou Dianzi University
Hangzhou Power Equipment Manufacturing Co Ltd
Hangzhou Power Supply Co of State Grid Zhejiang 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 Hangzhou Dianzi University, Hangzhou Power Equipment Manufacturing Co Ltd, Hangzhou Power Supply Co of State Grid Zhejiang Electric Power Co Ltd filed Critical Hangzhou Dianzi University
Priority to CN202110631295.2A priority Critical patent/CN113204926A/en
Publication of CN113204926A publication Critical patent/CN113204926A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • 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
    • 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/08—Probabilistic or stochastic CAD

Abstract

The application discloses an electric appliance prediction method, a device, equipment and a medium based on a scoring mechanism, wherein the method comprises the following steps: when the electric equipment to be tested with switching action in the target power system needs to be predicted, extracting target electrical characteristics of the electric equipment to be tested during switching action, respectively inputting the target electrical characteristics into all the two classification models in the two classification model set, and grading the probability that the electric equipment to be tested belongs to the positive samples and the negative samples in each two classification models by using a preset grading function; and then, carrying out accumulation statistics on scores of the electric equipment to be tested belonging to the same electric equipment, and judging the electric equipment corresponding to the highest accumulated score as the electric equipment to be tested. Because the method introduces a scoring mechanism into a plurality of binary models, the uniqueness and reliability of the prediction result of the electric appliance equipment to be tested can be ensured, and the accuracy of the electric appliance equipment to be tested in the prediction can be obviously improved.

Description

Electric appliance prediction method, device and equipment based on scoring mechanism
Technical Field
The invention relates to the technical field of power systems, in particular to an electrical equipment prediction method, device, equipment and medium based on a scoring mechanism.
Background
How to accurately predict the electrical equipment with switching action in the power system by a non-contact means becomes a current research hotspot. In the prior art, a one-to-many classification model is generally used to determine an electrical device in a power system, where a switching action occurs. That is, in the process of predicting the electrical equipment to be tested by using the one-to-many classification model, a plurality of two-classification models are trained for all electrical equipment in the power system, then, the two-classification models are used for performing prediction voting on the electrical equipment to be tested with switching actions, and finally, the electrical equipment with the highest voting result is used as the prediction result of the electrical equipment to be tested. However, the model prediction method may have a phenomenon that a plurality of highest votes are the same, in this case, one of the prediction results is randomly selected as a final prediction result of the electrical equipment to be tested, so that the prediction result of the electrical equipment to be tested has the problems of large error and low precision. At present, no effective solution exists for the technical problem.
Disclosure of Invention
In view of the above, the present invention provides a method, an apparatus, a device and a medium for predicting an electrical device based on a scoring mechanism, so as to improve the accuracy of the prediction of the electrical device to be tested. The specific scheme is as follows:
an electric appliance prediction method based on a scoring mechanism comprises the following steps:
when the electric equipment to be tested with switching action in a target power system needs to be predicted, extracting target electrical characteristics of the electric equipment to be tested when the electric equipment to be tested performs the switching action;
respectively inputting the target electrical characteristics into all the two classification models of the two classification model set, and grading the probability that the electrical equipment to be tested belongs to the positive sample and the negative sample in each two classification model by using a preset grading function to obtain a target grading set;
in the target scoring set, performing accumulation statistics on the scores of the electric equipment to be tested, which belong to the same electric equipment, and judging the electric equipment corresponding to the highest accumulated score as the electric equipment to be tested;
wherein the creation process of the two classification model sets comprises the following steps:
acquiring electrical characteristics of each electrical device in the target power system when the electrical device performs switching action;
and establishing a two-classification model according to the electrical characteristics of any two electrical devices in the target power system during switching action, so as to obtain the two-classification model set.
Preferably, the process of acquiring the electrical characteristics of each electrical device in the target power system during switching operation includes:
and acquiring the electrical characteristics of each electrical device in the target power system during switching action by using a non-invasive load monitoring technology.
Preferably, the process of obtaining the electrical characteristics of each electrical device in the target power system during the switching action by using the non-intrusive load monitoring technology includes:
acquiring the current amplitude of each electrical device in the target power system during switching action by using the non-invasive load monitoring technology;
and calculating the mean value and the mean square error of the current amplitude when each electrical device performs switching action, and respectively obtaining the electrical characteristics of each electrical device during switching action.
Preferably, the expression of the preset scoring function is as follows:
f(x)＝1/(1+e-x)；
in the formula, x is the Euclidean distance between the target electrical characteristic and a decision surface in a target two-classification model, and the target two-classification model is any one of the two-classification model sets.
Preferably, the process of establishing a two-class model according to electrical characteristics of any two electrical devices in the target power system during switching actions to obtain the two-class model set includes:
and establishing a two-classification model according to the electrical characteristics of any two electrical equipment in the target power system during switching action based on an SVM algorithm to obtain the two-classification model set.
Preferably, the process of establishing a two-classification model according to the electrical characteristics of any two electrical devices in the target power system during switching actions based on the SVM algorithm to obtain the two-classification model set includes:
and setting a kernel function in the SVM algorithm as a Gaussian function, and establishing a two-classification model according to the electrical characteristics of any two electrical equipment in the target power system during switching action to obtain the two-classification model set.
Correspondingly, the application also discloses an electrical equipment prediction device based on the scoring mechanism, which comprises:
the characteristic extraction module is used for extracting target electrical characteristics of the to-be-detected electrical equipment during switching action when the to-be-detected electrical equipment with switching action in a target power system needs to be predicted;
the scoring calculation module is used for respectively inputting the target electrical characteristics into all the two classification models in the two classification model set, and scoring the probability that the electrical equipment to be tested belongs to the positive sample and the negative sample in each two classification model by using a preset scoring function to obtain a target scoring set;
the electric appliance prediction module is used for performing accumulation statistics on the scores of the electric appliances to be tested belonging to the same electric appliance in the target score set and judging the electric appliance corresponding to the highest accumulated score as the electric appliance to be tested;
wherein the two classification model sets are created by a model creation module, and the model creation module comprises:
the characteristic acquisition unit is used for acquiring the electrical characteristics of each electrical device in the target power system during switching action;
and the model establishing unit is used for establishing a two-classification model according to the electrical characteristics of any two electrical equipment in the target power system during switching action to obtain the two-classification model set.
Correspondingly, the application also discloses an electrical equipment prediction device based on the scoring mechanism, which comprises:
a memory for storing a computer program;
a processor for implementing the steps of a scoring mechanism based appliance prediction method as disclosed in the foregoing when executing the computer program.
Accordingly, the present application also discloses a computer readable storage medium having a computer program stored thereon, which when executed by a processor implements the steps of a scoring mechanism based appliance prediction method as disclosed in the foregoing.
In the invention, in order to predict the electric equipment to be tested which has switching action in the target electric power system, firstly, the electric characteristics of each electric equipment in the target electric power system when switching action is carried out are obtained in advance, and a binary model is established according to the electric characteristics of any two electric equipment in the target electric power system when switching action is carried out, so as to obtain a binary model set; then, extracting target electrical characteristics of the electrical equipment to be tested during switching action, respectively inputting the target electrical characteristics into all the two classification models of the two classification model set, and simultaneously grading the probability of the electrical equipment to be tested belonging to a positive sample and a negative sample in each two classification model by using a preset grading function to obtain a target grading set; and finally, in the target score set, performing accumulation statistics on the scores of the electrical equipment to be tested belonging to the same electrical equipment, and judging the electrical equipment corresponding to the highest accumulated score as the electrical equipment to be tested. Obviously, compared with the prior art, the method introduces a scoring mechanism into the two classification models, so that the situation that the highest votes are the same when the electric equipment to be tested is predicted by using the one-to-many classification model in the prior art can be effectively avoided, the uniqueness and the reliability of the prediction result of the electric equipment to be tested are ensured, and the accuracy of the electric equipment to be tested in the prediction can be obviously improved. Correspondingly, the electric appliance prediction device and the electric appliance prediction medium based on the scoring mechanism have the beneficial effects.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a flowchart of an electrical device prediction method based on a scoring mechanism according to an embodiment of the present invention;
fig. 2 is a structural diagram of an electrical appliance prediction apparatus based on a scoring mechanism according to an embodiment of the present invention;
fig. 3 is a structural diagram of an electrical appliance prediction device based on a scoring mechanism according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, fig. 1 is a flowchart of an electrical device prediction method based on a scoring mechanism according to an embodiment of the present invention, where the prediction method includes:
step S11: when the electric equipment to be tested with switching action in the target power system needs to be predicted, extracting target electrical characteristics of the electric equipment to be tested when the switching action is carried out;
step S12: respectively inputting the target electrical characteristics into all the two classification models of the two classification model set, and grading the probability that the electrical equipment to be tested belongs to the positive sample and the negative sample in each two classification model by using a preset grading function to obtain a target grading set;
step S13: in the target scoring set, performing accumulation statistics on scores of the to-be-tested electrical equipment belonging to the same electrical equipment, and judging the electrical equipment corresponding to the highest accumulated score as the to-be-tested electrical equipment;
the creating process of the two classification model sets comprises the following steps:
acquiring electrical characteristics of each electrical device in a target power system during switching action;
and establishing a two-classification model according to the electrical characteristics of any two electrical equipment in the target power system during switching action to obtain a two-classification model set.
In this embodiment, an electrical device prediction method based on a scoring mechanism is provided to improve the accuracy of prediction of an electrical device to be tested. In the prediction method, a two-classification model set needs to be created in advance, in the process of creating the two-classification model set, firstly, the electrical characteristics of each electrical device in a target power system during switching action are obtained, and a two-classification model is established according to the electrical characteristics of any two electrical devices in the target power system during switching, so that the two-classification model set is obtained.
Assuming that there are n electrical devices in total in the target power system, s binary models are trained in the target power system. Wherein the content of the first and second substances,
Figure BDA0003103846500000051
in the formula, s is the number of the two classification models in the target power system, and n is the number of the electrical equipment in the target power system. Namely, randomly selecting 2 electrical equipment from n electrical equipment of the target power system to carry out combination training of the two-classification model until all the electrical equipment combinations in the target power system are traversed.
When the electrical equipment to be tested which has switching action in the target power system needs to be predicted, that is, when it needs to be determined which electrical equipment in the target power system has switching action, firstly, the target electrical characteristics of the electrical equipment to be tested when switching action is carried out are extracted; when the target electrical characteristics of the electrical equipment to be tested during switching action are obtained, the target electrical characteristics of the electrical equipment to be tested during switching action are respectively input into all the two classification models in the two classification model set, and the probability that the electrical equipment to be tested belongs to the positive samples and the negative samples in each two classification model is graded by using a preset grading function, so that a target grading set is obtained. And finally, accumulating and counting the scores of the electric equipment to be tested belonging to the same electric equipment in the target score set, and judging the electric equipment corresponding to the highest accumulated score as the electric equipment to be tested.
Here, a specific example is described, assuming that there are 3 electrical devices in the target power system, each of which is A, B, C, and at this time, there are 3 classification models in the two classification model sets, where the first classification model is a classification model of a and B, the second classification model is a classification model of a and C, and the third classification model is a classification model of B and C.
If the electrical equipment to be tested in the target power system needs to be predicted, the electrical characteristics of the electrical equipment to be tested need to be respectively input into the first classification model, the second classification model and the third classification model, and meanwhile, the probability that the electrical equipment to be tested belongs to the positive samples and the negative samples in the two classification models is scored by using a preset scoring function. That is, the predetermined scoring function is used to score the probabilities that the electrical equipment to be tested belongs to a and B in the first classification model, and similarly, the predetermined scoring function is used to score the probabilities that the electrical equipment to be tested belongs to a and C in the second classification model, and the predetermined scoring function is used to score the probabilities that the electrical equipment to be tested belongs to B and C in the third classification model.
It is assumed that the probabilities of the electrical equipment to be tested belonging to the first classification model and belonging to A and B are respectively 0.2 and 0.3, the probabilities of the electrical equipment to be tested belonging to the second classification model and belonging to A and C are respectively 0.3 and 0.8, and the probabilities of the electrical equipment to be tested belonging to the third classification model and belonging to B and C are respectively 0.3 and 0.8. Then the target score set is 0.2, 0.3, 0.8, 0.3, and 0.8. And then, accumulating and counting the scores of the electric appliances to be tested belonging to the same electric appliance. Specifically, in this embodiment, the probabilities that the electrical device to be tested belongs to the electrical device A, B, C are respectively obtained by statistics and are 0.5, 0.6, and 1.6, and in this case, the score that the electrical device to be tested belongs to C is the highest, which indicates that the electrical device to be tested is C. It should be noted that the electrical device under test may be any electrical device in the target power system, and is not specifically limited herein.
Compared with the prior art, the method introduces a scoring mechanism into the two-classification models, so that the condition that the highest votes are the same when the electric equipment to be tested is predicted by using the one-to-many classification models in the prior art can be effectively avoided, the uniqueness and the reliability of the prediction result of the electric equipment to be tested are ensured, and the accuracy of the electric equipment to be tested in the prediction process can be obviously improved.
In addition, if electrical equipment is newly added to the target power system, the purpose of predicting the electrical equipment to be tested can be achieved by using the prediction method in the prior art and retraining the classification model, and by the method provided by the embodiment, the purpose of predicting the electrical equipment to be tested can be achieved only by adding the classification model related to the newly added electrical equipment to the two classification model set. Obviously, the method provided by the embodiment can also relatively improve the convenience of people in using the method to predict the electrical equipment to be tested.
In this embodiment, in order to predict the electrical equipment to be tested, which has a switching action in the target electrical power system, first, electrical characteristics of each electrical equipment in the target electrical power system when the electrical equipment is in the switching action are obtained in advance, and a binary model is established according to the electrical characteristics of any two electrical equipment in the target electrical power system when the electrical equipment is in the switching action, so as to obtain a binary model set; then, extracting target electrical characteristics of the electrical equipment to be tested during switching action, respectively inputting the target electrical characteristics into all the two classification models of the two classification model set, and simultaneously grading the probability of the electrical equipment to be tested belonging to a positive sample and a negative sample in each two classification model by using a preset grading function to obtain a target grading set; and finally, in the target score set, performing accumulation statistics on the scores of the electrical equipment to be tested belonging to the same electrical equipment, and judging the electrical equipment corresponding to the highest accumulated score as the electrical equipment to be tested. Obviously, compared with the prior art, the method introduces a scoring mechanism into the two classification models, so that the situation that the highest votes are the same when the electric equipment to be tested is predicted by using the one-to-many classification model in the prior art can be effectively avoided, the uniqueness and the reliability of the prediction result of the electric equipment to be tested are ensured, and the accuracy of the electric equipment to be tested in the prediction can be obviously improved.
Based on the above embodiments, this embodiment further describes and optimizes the technical solution, and as a preferred implementation, the above steps: the process of obtaining the electrical characteristics of each electrical device in the target power system during switching action comprises the following steps:
and acquiring the electrical characteristics of each electrical device in the target power system during switching action by using a non-invasive load monitoring technology.
Specifically, in this embodiment, a Non-Intrusive Load Monitoring (NILM) technology is used to obtain electrical characteristics of each electrical device in the target power system during switching operation. The technology can be used for installing monitoring equipment at a power inlet, and the type and the operation condition of a single load in the load cluster can be analyzed and obtained by monitoring a voltage and current signal at a specific position.
Therefore, through the technical scheme provided by the embodiment, the extraction process of the electrical characteristics can be simpler and more reliable, and the manpower and time investment required by installation and maintenance of the monitoring equipment can be saved.
As a preferred embodiment, the above steps: the process of acquiring the electrical characteristics of each electrical device in the target power system during switching action by using the non-intrusive load monitoring technology comprises the following steps:
acquiring the current amplitude of each electrical device in a target power system during switching action by using a non-invasive load monitoring technology;
and calculating the mean value and the mean square error of the current amplitude when each electrical device performs switching action, and respectively obtaining the electrical characteristics of each electrical device during switching action.
In this embodiment, a specific operation method for extracting electrical characteristics of each electrical device in a target power system is provided, that is, in a process of extracting electrical characteristics of each electrical device in a switching operation of the target power system, first, a current amplitude of each electrical device in the target power system in the switching operation is obtained by using a non-intrusive load monitoring technology, and then, a mean value and a mean square deviation of the current amplitudes of each electrical device in the switching operation are calculated, so that the electrical characteristics of each electrical device in the switching operation can be obtained.
The measurement process is relatively simple and easy compared with other electrical parameters of the electrical equipment, so that when the electrical characteristics of the electrical equipment are extracted in the switching action by the arrangement mode, convenience in extracting the electrical characteristics of the electrical equipment can be relatively improved.
Based on the above embodiments, this embodiment further describes and optimizes the technical solution, and as a preferred implementation, the expression of the preset scoring function is as follows:
f(x)＝1/(1+e-x)；
in the formula, x is the Euclidean distance between the target electrical characteristic and a decision surface in a target two-classification model, and the target two-classification model is any one two-classification model in a two-classification model set.
Specifically, in the present embodiment, a preset scoring function f (x) is used, where f (x) is 1/(1+ e)-x) Come to rightAnd scoring the probability of the electrical equipment to be tested belonging to the positive sample and the negative sample in each two-classification model. And x is the Euclidean distance between the target electrical characteristic and the decision surface in the target two-class model, namely, the mean value and the mean square error of the current amplitude of the electric equipment to be tested are used as the Euclidean distance between the coordinate point and the decision surface in each two-class model when the electric equipment to be tested performs switching action.
Specifically, if the mean value and the mean square deviation of the current amplitude of the electrical equipment to be tested are taken as the positions of the coordinate points and are located in the positive sample area of the target two-classification model, the higher the score number of the preset scoring function f (x) is, the higher the probability that the electrical equipment to be tested belongs to the positive sample in the target two-classification model is, the higher the probability that the electrical equipment to be tested belongs to the negative sample in the target two-classification model is, and the lower the probability that the electrical equipment to be tested belongs to the negative sample in the target two-classification model is; if the mean value and the mean square deviation of the current amplitude of the electrical equipment to be tested are taken as the positions of the coordinate points and are located in the negative sample area of the target two-classification model, the higher the score number of the preset scoring function f (x) is, the higher the probability that the electrical equipment to be tested belongs to the negative sample in the target two-classification model is, and the lower the probability that the electrical equipment to be tested belongs to the positive sample in the target two-classification model is.
Based on the above embodiments, this embodiment further describes and optimizes the technical solution, and as a preferred implementation, the above steps: the process of establishing a two-classification model according to the electrical characteristics of any two electrical devices in a target power system during switching action to obtain a two-classification model set comprises the following steps:
based on an SVM algorithm, establishing a two-classification model according to the electrical characteristics of any two electrical equipment in the target power system during switching action, and obtaining a two-classification model set.
It can be understood that, because the SVM (Support Vector Machine) algorithm not only opens the source of the algorithm code, but also has a good classification effect. Therefore, in this embodiment, a binary model set is obtained by establishing a binary model according to the electrical characteristics of any two electrical devices in the target power system during switching operation based on the SVM algorithm.
As a preferred embodiment, the above steps: the process of establishing a two-classification model according to the electrical characteristics of any two electrical devices in a target power system during switching action based on an SVM algorithm to obtain a two-classification model set comprises the following steps:
setting a kernel function in the SVM algorithm as a Gaussian function, and establishing a two-classification model according to the electrical characteristics of any two electrical equipment in the target power system during switching action to obtain a two-classification model set.
Specifically, in this embodiment, a kernel function in the SVM algorithm is set as a gaussian function to establish a binary model. After the kernel function in the SVM algorithm is set to be the Gaussian function, the problem of nonlinear classification of the two-classification model can be solved, and the Gaussian function has only one parameter compared with other kernel functions of the SVM, so that the selection is easy.
Therefore, the technical scheme provided by the embodiment can further reduce the difficulty in creating the two-classification model.
Referring to fig. 2, fig. 2 is a structural diagram of an electrical appliance prediction apparatus based on a scoring mechanism according to an embodiment of the present invention, where the prediction apparatus includes:
the feature extraction module 21 is configured to, when it is required to predict an electrical device to be tested that performs a switching action in a target power system, extract a target electrical feature of the electrical device to be tested during the switching action;
the scoring calculation module 22 is used for respectively inputting the target electrical characteristics into all the two classification models in the two classification model set, and scoring the probability that the electrical equipment to be tested belongs to the positive sample and the negative sample in each two classification model by using a preset scoring function to obtain a target scoring set;
the electrical appliance prediction module 23 is configured to perform cumulative statistics on scores of the electrical appliances to be tested belonging to the same electrical appliance in the target score set, and determine the electrical appliance corresponding to the highest cumulative score as the electrical appliance to be tested;
the two-classification model set is created by a model creating module, and the model creating module comprises:
the characteristic acquisition unit is used for acquiring the electrical characteristics of each electrical device in the target power system during switching action;
and the model establishing unit is used for establishing a two-classification model according to the electrical characteristics of any two electrical equipment in the target power system during switching action to obtain a two-classification model set.
The electric appliance prediction device based on the scoring mechanism has the beneficial effects of the electric appliance prediction method based on the scoring mechanism.
Referring to fig. 3, fig. 3 is a structural diagram of an electrical device prediction device based on a scoring mechanism according to an embodiment of the present invention, where the prediction device includes:
a memory 31 for storing a computer program;
a processor 32 for implementing the steps of a scoring mechanism based appliance prediction method as disclosed in the foregoing when executing the computer program.
The electric appliance prediction device based on the scoring mechanism has the beneficial effects of the electric appliance prediction method based on the scoring mechanism.
Correspondingly, the embodiment of the invention also discloses a computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, and when the computer program is executed by a processor, the steps of the electric appliance prediction method based on the scoring mechanism are realized.
The computer-readable storage medium provided by the embodiment of the invention has the beneficial effects of the electric appliance prediction method based on the scoring mechanism.
The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The above describes in detail an electrical equipment prediction method, apparatus, device and medium based on a scoring mechanism, which are provided by the present invention, and a specific example is applied in the present document to explain the principle and implementation manner of the present invention, and the description of the above embodiment is only used to help understanding the method and core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.

Claims (9)

1. An electric appliance prediction method based on a scoring mechanism is characterized by comprising the following steps:
when the electric equipment to be tested with switching action in a target power system needs to be predicted, extracting target electrical characteristics of the electric equipment to be tested when the electric equipment to be tested performs the switching action;
respectively inputting the target electrical characteristics into all the two classification models of the two classification model set, and grading the probability that the electrical equipment to be tested belongs to the positive sample and the negative sample in each two classification model by using a preset grading function to obtain a target grading set;
in the target scoring set, performing accumulation statistics on the scores of the electric equipment to be tested, which belong to the same electric equipment, and judging the electric equipment corresponding to the highest accumulated score as the electric equipment to be tested;
wherein the creation process of the two classification model sets comprises the following steps:
acquiring electrical characteristics of each electrical device in the target power system when the electrical device performs switching action;
and establishing a two-classification model according to the electrical characteristics of any two electrical devices in the target power system during switching action, so as to obtain the two-classification model set.
2. The electrical equipment prediction method according to claim 1, wherein the process of obtaining the electrical characteristics of each electrical equipment in the target power system during switching operation includes:
and acquiring the electrical characteristics of each electrical device in the target power system during switching action by using a non-invasive load monitoring technology.
3. The electrical equipment prediction method according to claim 2, wherein the process of obtaining the electrical characteristics of each electrical equipment in the target power system during switching by using a non-intrusive load monitoring technology comprises:
acquiring the current amplitude of each electrical device in the target power system during switching action by using the non-invasive load monitoring technology;
and calculating the mean value and the mean square error of the current amplitude when each electrical device performs switching action, and respectively obtaining the electrical characteristics of each electrical device during switching action.
4. The electrical equipment prediction method according to claim 1, wherein the expression of the preset scoring function is:
f(x)＝1/(1+e-x)；
in the formula, x is the Euclidean distance between the target electrical characteristic and a decision surface in a target two-classification model, and the target two-classification model is any one of the two-classification model sets.
5. The electrical equipment prediction method according to any one of claims 1 to 4, wherein the process of establishing a two-classification model according to electrical characteristics of any two electrical equipment in the target power system during switching operation to obtain the two-classification model set includes:
and establishing a two-classification model according to the electrical characteristics of any two electrical equipment in the target power system during switching action based on an SVM algorithm to obtain the two-classification model set.
6. The electrical equipment prediction method according to claim 5, wherein the process of establishing a two-classification model according to electrical characteristics of any two electrical equipment in the target power system during switching actions based on the SVM algorithm to obtain the two-classification model set comprises:
and setting a kernel function in the SVM algorithm as a Gaussian function, and establishing a two-classification model according to the electrical characteristics of any two electrical equipment in the target power system during switching action to obtain the two-classification model set.
7. An electrical equipment prediction device based on a scoring mechanism is characterized by comprising:
the characteristic extraction module is used for extracting target electrical characteristics of the to-be-detected electrical equipment during switching action when the to-be-detected electrical equipment with switching action in a target power system needs to be predicted;
the scoring calculation module is used for respectively inputting the target electrical characteristics into all the two classification models in the two classification model set, and scoring the probability that the electrical equipment to be tested belongs to the positive sample and the negative sample in each two classification model by using a preset scoring function to obtain a target scoring set;
the electric appliance prediction module is used for performing accumulation statistics on the scores of the electric appliances to be tested belonging to the same electric appliance in the target score set and judging the electric appliance corresponding to the highest accumulated score as the electric appliance to be tested;
wherein the two classification model sets are created by a model creation module, and the model creation module comprises:
the characteristic acquisition unit is used for acquiring the electrical characteristics of each electrical device in the target power system during switching action;
and the model establishing unit is used for establishing a two-classification model according to the electrical characteristics of any two electrical equipment in the target power system during switching action to obtain the two-classification model set.
8. An electrical equipment prediction device based on a scoring mechanism, comprising:
a memory for storing a computer program;
a processor for implementing the steps of a scoring mechanism based appliance prediction method as claimed in any one of claims 1 to 6 when executing said computer program.
9. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when being executed by a processor, carries out the steps of a scoring mechanism-based appliance prediction method according to any one of claims 1 to 6.
CN202110631295.2A 2021-06-07 2021-06-07 Electric appliance prediction method, device and equipment based on scoring mechanism Pending CN113204926A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110631295.2A CN113204926A (en) 2021-06-07 2021-06-07 Electric appliance prediction method, device and equipment based on scoring mechanism

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110631295.2A CN113204926A (en) 2021-06-07 2021-06-07 Electric appliance prediction method, device and equipment based on scoring mechanism

Publications (1)

Publication Number Publication Date
CN113204926A true CN113204926A (en) 2021-08-03

Family

ID=77024125

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110631295.2A Pending CN113204926A (en) 2021-06-07 2021-06-07 Electric appliance prediction method, device and equipment based on scoring mechanism

Country Status (1)

Country Link
CN (1) CN113204926A (en)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111444930A (en) * 2019-01-17 2020-07-24 上海游昆信息技术有限公司 Method and device for determining prediction effect of two-classification model

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111444930A (en) * 2019-01-17 2020-07-24 上海游昆信息技术有限公司 Method and device for determining prediction effect of two-classification model

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
吕干云等: "一种基于二分类的电压暂降源定位方法", 《电力系统保护与控制》 *
杨珮鑫 等: "基于多分类支持向量机的分布式发电系统并网保护", 《电力系统自动化》 *
程玉虎等: "基于概率型支持向量分类机的Q学习", 《中国矿业大学学报》 *

Similar Documents

Publication Publication Date Title
CN106940679B (en) Data processing method and device
CN112699913A (en) Transformer area household variable relation abnormity diagnosis method and device
CN110007220B (en) Method and device for diagnosing operating state of circuit breaker mechanism
CN110533089B (en) Self-adaptive non-invasive load identification method based on random forest
CN116559598B (en) Smart distribution network fault positioning method and system
CN113340353B (en) Monitoring method, equipment and medium for power transmission line
CN112990330A (en) User energy abnormal data detection method and device
CN116739829B (en) Big data-based power data analysis method, system and medium
CN110751385A (en) Non-invasive load identification method, terminal device and storage medium
CN112036450B (en) High-voltage cable partial discharge mode identification method and system based on transfer learning
CN114240817A (en) Data analysis method and device, electronic equipment and storage medium
CN110348717B (en) Base station value scoring method and device based on grid granularity
CN116401532A (en) Method and system for recognizing frequency instability of power system after disturbance
CN111881159A (en) Fault detection method and device based on cost-sensitive extreme random forest
Ouyang et al. Optimisation of time window size for wind power ramps prediction
EP3285206B1 (en) Evaluation device, evaluation method and evaluation program for biometric data
CN117289167A (en) Battery remaining life prediction method, device and medium based on multiple neural network
CN113204926A (en) Electric appliance prediction method, device and equipment based on scoring mechanism
CN110703013B (en) Online identification method and device for low-frequency oscillation mode of power system and electronic equipment
CN117236022A (en) Training method and application method of residual life prediction model of transformer and electronic equipment
CN110704614B (en) Information processing method and device for predicting user group type in application
CN111506636A (en) System and method for analyzing residential electricity consumption behavior based on autoregressive and neighbor algorithm
CN115408927A (en) Data processing method and device for predicting rock mass parameters
CN114414951A (en) Method and system for diagnosing insulation defect of gas insulated metal enclosed switchgear
CN115422821A (en) Data processing method and device for rock mass parameter prediction

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20210803

RJ01 Rejection of invention patent application after publication