CN112632857A - Method, device, equipment and storage medium for determining line loss of power distribution network - Google Patents
Method, device, equipment and storage medium for determining line loss of power distribution network Download PDFInfo
- Publication number
- CN112632857A CN112632857A CN202011531544.2A CN202011531544A CN112632857A CN 112632857 A CN112632857 A CN 112632857A CN 202011531544 A CN202011531544 A CN 202011531544A CN 112632857 A CN112632857 A CN 112632857A
- Authority
- CN
- China
- Prior art keywords
- line data
- sample
- detected
- support vector
- vector regression
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000009826 distribution Methods 0.000 title claims abstract description 45
- 238000003860 storage Methods 0.000 title claims abstract description 9
- 239000013598 vector Substances 0.000 claims abstract description 93
- 238000012549 training Methods 0.000 claims abstract description 53
- 238000004364 calculation method Methods 0.000 claims abstract description 24
- 238000012545 processing Methods 0.000 claims abstract description 23
- 230000008569 process Effects 0.000 claims abstract description 15
- 238000004590 computer program Methods 0.000 claims description 10
- 238000000638 solvent extraction Methods 0.000 claims 1
- 238000004422 calculation algorithm Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 101100517651 Caenorhabditis elegans num-1 gene Proteins 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
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
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/04—Constraint-based CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/04—Power grid distribution networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a method, a device, equipment and a storage medium for determining line loss of a power distribution network, wherein the method comprises the following steps: generating a plurality of subclasses and a plurality of support vector regression machines to be selected through a preset support vector regression machine training process, acquiring line data from a power distribution network, and carrying out standardization processing on the line data to generate line data to be detected; classifying the line data to be detected based on the plurality of subclasses to obtain the class of the line data to be detected; selecting a target support vector regression machine from a plurality of support vector regression machines to be selected according to the type of the line data to be detected; and inputting the line data to be detected into a target support vector regression machine, and determining the line loss corresponding to the line data to be detected. The method can effectively improve the efficiency and accuracy of line loss calculation.
Description
Technical Field
The present invention relates to the field of line loss calculation technologies, and in particular, to a method, an apparatus, a device, and a storage medium for determining line loss of a power distribution network.
Background
Under the new period environment, as people have greater demand and dependence on electric energy, higher requirements are put on a power supply system, and related power enterprises pay great attention to the economic development of electric power in order to better realize the provision of power supply services. The line loss, which is a main part of the power loss in the power distribution network, is always a difficult problem to be solved and explored by the power system.
Line loss, line loss for short, is the energy loss generated by the transmission of electrical energy through a transmission line.
At present, a plurality of methods exist for calculating line loss, for example, traditional algorithms include a root mean square current method, a newton method, a linear regression analysis method and the like, but in practice, data are often lost, nodes of a power network are more, and operating data and structural parameters of a power grid have a difficult situation of being difficult to collect and arrange, so that the line loss calculation efficiency is low, and the actual line loss situation in a power distribution network is difficult to accurately determine.
Disclosure of Invention
The invention provides a method, a device, equipment and a storage medium for determining line loss of a power distribution network, and solves the technical problems of low line loss calculation efficiency and low accuracy rate caused by difficulty in quickly and accurately collecting power grid operation data due to data loss or more power grid nodes in the prior art.
The invention provides a method for determining line loss of a power distribution network, which comprises the following steps:
acquiring line data from a power distribution network, and carrying out standardized processing on the line data to generate line data to be detected;
classifying the line data to be detected based on a plurality of subclasses to obtain the class of the line data to be detected;
selecting a target support vector regression machine from a plurality of support vector regression machines to be selected according to the category of the line data to be detected; the plurality of subclasses and the plurality of support vector regression machines to be selected are generated through a preset support vector regression machine training process;
and inputting the line data to be detected into the target support vector regression machine, and determining the line loss corresponding to the line data to be detected.
Optionally, the training process of the support vector regression machine includes:
acquiring a training sample set;
carrying out standardization processing on the training samples in the training sample set to generate a standardized sample set; the normalized sample set includes a plurality of normalized samples;
classifying the standardized sample set by adopting a preset classifier to obtain a plurality of subclasses;
and respectively training a plurality of preset initial support vector regression machines by adopting standardized samples in a plurality of subclasses to obtain a plurality of support vector regression machines to be selected.
Optionally, the step of classifying the normalized sample set by using a preset classifier to obtain a plurality of subclasses includes:
selecting a first target normalized sample from the normalized sample set as a first centroid sample using a preset classifier;
calculating a first Euclidean distance between each of the normalized samples and the first centroid sample;
classifying the category of the standardized samples with the first Euclidean distance smaller than or equal to a first preset clustering radius into a subclass and deleting the subclass from the standardized sample set;
selecting a second target normalized sample as a second centroid sample from the normalized samples having the first Euclidean distance greater than the first preset clustering radius;
updating the first centroid sample with the second centroid sample, returning to the step of calculating the first euclidean distance between each of the normalized samples and the first centroid sample until all of the normalized samples have a corresponding subclass.
Optionally, the plurality of subclasses respectively have corresponding class centroid samples, and the step of classifying the line data to be detected based on the plurality of subclasses to obtain the class of the line data to be detected includes:
respectively calculating a second Euclidean distance between the line data to be detected and each class centroid sample;
and taking the subclass corresponding to the class centroid sample with the second Euclidean distance smaller than or equal to a second preset clustering radius as the class of the line data to be detected.
The invention also provides a device for determining the line loss of the power distribution network, which comprises the following components:
the line data acquisition module to be detected is used for acquiring line data from the power distribution network, and carrying out standardized processing on the line data to generate line data to be detected;
the category determining module is used for classifying the line data to be detected based on a plurality of subclasses to obtain the category of the line data to be detected;
the target support vector regression selection module is used for selecting a target support vector regression from a plurality of support vector regression to be selected according to the type of the line data to be detected; the plurality of subclasses and the plurality of support vector regression machines to be selected are generated through a preset support vector regression machine training module;
and the line loss calculation module is used for inputting the line data to be detected into the target support vector regression machine and determining the line loss corresponding to the line data to be detected.
Optionally, the support vector regression training module includes:
the training sample set acquisition submodule is used for acquiring a training sample set;
the standardized sample set generation submodule is used for carrying out standardized processing on the training samples in the training sample set to generate a standardized sample set; the normalized sample set includes a plurality of normalized samples;
the subclass division submodule is used for classifying the standardized sample set by adopting a preset classifier to obtain a plurality of subclasses;
and the support vector regression training submodule is used for training a plurality of preset initial support vector regression machines by adopting standardized samples in a plurality of subclasses to obtain a plurality of support vector regression machines to be selected.
Optionally, the sub-classification sub-module includes:
a first centroid sample selection unit, configured to select a first target normalized sample from the normalized sample set as a first centroid sample using a preset classifier;
a first euclidean distance calculating unit for calculating a first euclidean distance between each of the normalized samples and the first centroid sample;
a classification unit, configured to classify a class of the normalized samples, for which the first euclidean distance is less than or equal to a first preset clustering radius, into a subclass and delete the subclass from the set of normalized samples;
a second centroid sample selection unit, configured to select a second target normalized sample from normalized samples whose first euclidean distance is greater than the first preset clustering radius as a second centroid sample;
a loop unit, configured to update the first centroid sample with the second centroid sample, and return to the step of calculating the first euclidean distance between each normalized sample and the first centroid sample until all normalized samples have a corresponding subclass.
Optionally, the plurality of sub-classes respectively have corresponding class centroid samples, and the class determination module includes:
the second Euclidean distance calculation submodule is used for calculating second Euclidean distances between the line data to be detected and each class centroid sample respectively;
and the class determination submodule is used for taking the subclass corresponding to the class centroid sample of which the second Euclidean distance is smaller than or equal to a second preset clustering radius as the class of the line data to be detected.
The present invention also provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method for determining line loss of a power distribution network according to any one of the above.
The invention also provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by the processor, implements the line loss determination method of the power distribution network according to any one of the above.
According to the technical scheme, the invention has the following advantages:
the method comprises the steps of obtaining line data from a power distribution network, carrying out standardized processing on the line data to generate line data to be detected, classifying the line data to be detected according to a plurality of preset subclasses, and determining the category of the line data to be detected; and selecting a target support vector regression machine from a plurality of pre-trained support vector regression machines to be selected based on the type of the line data to be detected, and finally calculating the line data to be detected by using the target support vector regression machine to determine the corresponding line loss. Therefore, the technical problems of low line loss calculation efficiency and low accuracy caused by the fact that the power grid operation data are difficult to collect quickly and accurately due to data loss or more power grid nodes in the prior art are solved, and the line loss calculation efficiency and accuracy are effectively improved.
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, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without inventive exercise.
Fig. 1 is a flowchart illustrating steps of a method for determining line loss of a power distribution network according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating steps of a method for determining line loss of a power distribution network according to an alternative embodiment of the present invention;
fig. 3 is a block diagram of a line loss determination apparatus for a power distribution network according to an embodiment of the present invention.
Detailed Description
The embodiment of the invention provides a method, a device, equipment and a storage medium for determining line loss of a power distribution network, which are used for solving the technical problems of low line loss calculation efficiency and low accuracy rate caused by difficulty in quickly and accurately collecting power grid operation data due to data loss or more power grid nodes in the prior art.
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the embodiments described below 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 illustrating a method for determining line loss of a power distribution network according to an embodiment of the present invention.
The invention provides a method for determining line loss of a power distribution network, which comprises the following steps:
in the specific implementation, parameters affecting the line loss of the power distribution network are various, line data can be obtained after actual analysis of the power distribution network, different dimensions may exist in the usually obtained line data, and in order to avoid the influence of the dimensions on the subsequent calculation process, the line data can be subjected to standardization processing to obtain line data to be detected, and the line data to be detected is used as the input quantity of the subsequent calculation of the line loss of the power distribution network.
It should be noted that the parameters affecting the line loss of the power distribution network may be power supply amounts of active power and reactive power of a line, lengths of the line, total capacity of transformers, the number of transformers, total number of line interruptions, current flowing through, and parameters obtained by an automatic meter in the line, and the like, which are not limited in this embodiment of the present invention.
102, classifying the line data to be detected based on a plurality of subclasses to obtain the class of the line data to be detected;
in the embodiment of the invention, the plurality of subclasses can be generated through a preset support vector regression machine training process, after the line data to be detected is obtained, the line data to be detected can be classified according to a plurality of subclasses obtained in the training process as a standard due to the influence of different regions, different industries or other factors, so that the data to be detected is classified into corresponding classes, and the line loss of the line data to be detected can be determined more accurately.
103, selecting a target support vector regression machine from a plurality of support vector regression machines to be selected according to the type of the line data to be detected;
in the embodiment of the present invention, the plurality of subclasses and the plurality of to-be-selected support vector regression machines are generated through a preset support vector regression machine training process, each to-be-selected support vector regression machine corresponds to one subclass, and the category of the line data to be detected is divided according to the plurality of subclasses. Therefore, after the category of the line data to be detected is obtained, a target support vector regression machine can be selected from the support vector regression machines to be selected according to the category of the line data to be detected, and line loss calculation is prepared.
And 104, inputting the line data to be detected into the target support vector regression machine, and determining the line loss corresponding to the line data to be detected.
A Support Vector Regression (SVR) is used for converting nonlinear transformation of an actual problem into a high-dimensional feature space, linear Regression is realized by constructing a linear decision function in the high-dimensional space after dimension increasing, different kernel functions are selected, and different SVM can be generated.
In a specific implementation, the line loss corresponding to the line data to be detected can be determined by inputting the line data to be detected into a target support vector regression machine.
In the embodiment of the invention, the line data are acquired from the power distribution network and are subjected to standardization processing to generate the line data to be detected, the line data to be detected are classified according to a plurality of preset subclasses, and the category of the line data to be detected is determined; and selecting a target support vector regression machine from a plurality of pre-trained support vector regression machines to be selected based on the type of the line data to be detected, and finally calculating the line data to be detected by using the target support vector regression machine to determine the corresponding line loss. Therefore, the technical problems of low line loss calculation efficiency and low accuracy caused by the fact that the power grid operation data are difficult to collect quickly and accurately due to data loss or more power grid nodes in the prior art are solved, and the line loss calculation efficiency and accuracy are effectively improved.
Referring to fig. 2, fig. 2 is a flowchart illustrating a method for determining line loss of a power distribution network according to an alternative embodiment of the present invention.
The invention provides a method for determining line loss of a power distribution network, which comprises the following steps:
in a specific implementation, the parameters most relevant to the line loss of the power distribution network may be a power supply value of active power of a line, a power supply value of reactive power of the line, a total capacity of a transformer and a total length value of the line, so that the parameters can be matched through matchingMeasuring attribute values of each node in a line of the power grid, difference values between adjacent nodes, data of a meter and the like to obtain the four quantity characteristics as line data X to be detected, and using a multidimensional array (X)1，x2，x3，x4) Is expressed in terms of the form.
It should be noted that the process of the normalization processing may be specifically configured and processed according to a specific dimension of the line data to be detected, for example, normalization processing, and the like, which is not limited in this embodiment of the present invention.
Optionally, the plurality of subclasses respectively have corresponding class centroid samples, and the step 102 may be replaced by the following step 202 and step 203:
in the embodiment of the invention, each subclass is provided with a corresponding class centroid sample, and the similarity between the line data to be detected and the subclass corresponding to the class centroid sample can be determined by calculating the second Euclidean distance between the line data to be detected and each class centroid sample.
in an optional embodiment of the present invention, according to a comparison result between a second euclidean distance between the line data to be detected and each category centroid sample and a second preset clustering radius, if the comparison result is less than or equal to the second preset clustering radius, the subclass corresponding to the category centroid sample is used as the category of the line data to be detected.
204, selecting a target support vector regression machine from a plurality of support vector regression machines to be selected according to the type of the line data to be detected;
the plurality of subclasses and the plurality of support vector regression machines to be selected are generated through a preset support vector regression machine training process;
further, the support vector regression training process includes the following steps S1-S4:
s1, obtaining a training sample set;
s2, carrying out standardization processing on the training samples in the training sample set to generate a standardized sample set;
the normalized sample set includes a plurality of normalized samples;
s3, classifying the standardized sample set by adopting a preset classifier to obtain a plurality of subclasses;
in one embodiment of the present invention, the step S3 may include the following sub-steps:
selecting a first target normalized sample from the normalized sample set as a first centroid sample using a preset classifier;
calculating a first Euclidean distance between each of the normalized samples and the first centroid sample;
classifying the category of the standardized samples with the first Euclidean distance smaller than or equal to a first preset clustering radius into a subclass and deleting the subclass from the standardized sample set;
selecting a second target normalized sample as a second centroid sample from the normalized samples having the first Euclidean distance greater than the first preset clustering radius;
updating the first centroid sample with the second centroid sample, returning to the step of calculating the first euclidean distance between each of the normalized samples and the first centroid sample until all of the normalized samples have a corresponding subclass.
In the embodiment of the invention, a preset classifier is adopted to select a first target standardized sample from a standardized sample set as a first centroid sample, then a first Euclidean distance between each standardized sample and the first centroid sample is calculated, if the first Euclidean distance is less than or equal to a first preset clustering radius, the standardized sample is classified into a subclass, and the classified standardized sample is deleted from the standardized sample set; if the first euclidean distance is greater than the first preset clustering radius, then a second target normalized sample is selected from the normalized samples as a second centroid sample in preparation for creating a new subclass, the first centroid sample may be updated with the second centroid sample, and the above loop is repeated until all normalized samples are classified into the corresponding subclass.
In a specific implementation, the number of clusters is assumed to be C, and a simple unsupervised clustering algorithm is used to perform clustering processing on the C-type data. The data can be clustered according to the preset clustering radius, and the clustering speed is higher than that of the traditional K-means algorithm.
For normalized sample set X ═ X1，x2，...，xNThe normalized sample set of, assuming its first preset clustering radius is R, the description of this algorithm is: c1＝{x1}, first centroid sample O1＝x1Class number cluster _ num 1, Z { x ═ x1，x2，...，xN}; selecting x ∈ Z sample, finding the existing centroid (i.e. first centroid sample O)j) Calculating the Euclidean distance d (x) between each element and the first centroid samplei，Oj) (ii) a If d (x)i，Oj) If R is less than or equal to R, x is addediAdding to class CjIn (i.e. C)j＝Cj+{xi}, class CjThe element in (1) is { c1，c1，…，cN}; if d (x)i，Oj)>R, a new class is added as cluster _ num + l, and Ccluster_num＝{xi}，Ocluster_num＝xiWhen Z is Z- { xi}; and stopping clustering if the set element is zero.
And S4, training a plurality of preset initial support vector regression machines by adopting the standardized samples in the subclasses respectively to obtain a plurality of support vector regression machines to be selected.
In the embodiment of the present invention, the normalized sample obtained by performing the sample classification process to obtain a certain class of training is { (x, z)i) 1, 2, N, the trained samples are fitted by a spatially linear function of high-dimensional features, this functionThe number is denoted as f (x) W · Φ (x) + b, and Φ (-) maps the trained data from the input space to the space of the high-dimensional feature, thereby transforming the fitting problem of non-linearity in the input space into the problem of spatial linear fitting of the high-dimensional feature. Considering the real-time requirement of the negotiation problem, in the model of this embodiment, the sample is learned by the regression engine of the support vector, and then the regression problem of the algorithm in the regression engine of the support vector can be expressed as the problem of constrained optimization, and the unconstrained optimization problem is transformed by the Lagrange method, and then the structure of the learning sample is combined to obtain the specific output of the regression engine of the support vector, that is, the fitting function, expressed as:
the fitting function represents a specific line loss calculation function of the power distribution network, a sample x to be calculated is input into the fitting function, the obtained output is a line loss value, training of an initial support vector regression machine is carried out by the method until the output line loss value is consistent with an actual line loss value corresponding to a standardized sample, the training is successful, and a plurality of support vector machines to be selected are obtained.
After the target support vector machine is obtained, the line loss corresponding to the line data to be detected is determined by inputting the line data to be detected into the target support vector regression machine.
In specific implementation, after the line loss value of the line is obtained, in order to verify the superiority and inferiority of the model, verification may be performed in an experimental verification manner, for example:
the data of the corresponding laboratory is selected, the line loss of 68 lines is accurately calculated, the validity and the practicability of the model are verified, the embodiment trains and analyzes the data of the line loss samples in the 68 groups, 58 groups are used as the training samples, and the other 10 groups are used as the testing samples.
The sample data of the 58 lines are classified by an unsupervised clustering algorithm, the embodiment mainly divides the samples into 4 types, the regression machine training of the support vectors is respectively carried out on the data of the 4 types of samples, and the regression simulator of the corresponding 4 types of support vectors can be obtained after the training is finished. Classifying 10 groups of test sample related data, simulating by regression machines SVR of respective support vectors, and outputting to obtain the following result
The results of Table 1.
| Sample number | Actual line loss | Simulated line loss | Absolute error | Relative error |
| 1 | 3.13 | 3.155 | 0.034 | 1.09％ |
| 2 | 5.97 | 6.117 | 0.156 | 2.61％ |
| 3 | 1.05 | 1.099 | 0.054 | 5.14％ |
| 4 | 2.68 | 2.687 | 0.026 | 0.97％ |
| 5 | 3.76 | 3.703 | 0.050 | 1.33％ |
| 6 | 3.39 | 3.440 | 0.153 | 4.51％ |
| 7 | 6.22 | 6.293 | 0.081 | 1.30％ |
| 8 | 0.98 | 0.934 | 0.039 | 3.98％ |
| 9 | 3.44 | 3.525 | 0.094 | 2.73％ |
| 10 | 4.58 | 4.349 | 0.224 | 4.89％ |
TABLE 1
As can be seen from the table, the regression simulator of the support vector has a high-precision line loss calculation result, and the error is low.
The effectiveness of the model algorithm of the embodiment is further verified, and 1 regression simulator of the support vectors can be obtained by directly carrying out processing through the regression simulator of the support vectors without carrying out classification on the trained samples; then, 10 sets of test sample related data are output through simulation calculation implemented by the regression simulator of the support vector, and the results shown in the following table 2 are obtained:
TABLE 2
By comparing the results in tables 1 and 2, it can be seen that the accuracy of the obtained results is relatively low and the average relative error thereof exceeds 10% in the calculation of the line loss to the power distribution network by the regression simulator that does not pass the classified preprocessed support vector, as compared with the regression simulator that passes the classified preprocessed support vector. Therefore, the training samples are subjected to classification preprocessing, and the efficiency of calculating the loss model can be effectively improved.
In the embodiment of the invention, the line data are acquired from the power distribution network and are subjected to standardization processing to generate the line data to be detected, the line data to be detected are classified according to a plurality of preset subclasses, and the category of the line data to be detected is determined; and selecting a target support vector regression machine from a plurality of pre-trained support vector regression machines to be selected based on the type of the line data to be detected, and finally calculating the line data to be detected by using the target support vector regression machine to determine the corresponding line loss. Therefore, the technical problems of low line loss calculation efficiency and low accuracy caused by the fact that the power grid operation data are difficult to collect quickly and accurately due to data loss or more power grid nodes in the prior art are solved, and the line loss calculation efficiency and accuracy are effectively improved.
Referring to fig. 3, fig. 3 is a block diagram illustrating a line loss determining apparatus of a power distribution network according to an embodiment of the present invention.
The invention provides a line loss determination device of a power distribution network, which comprises:
the to-be-detected line data acquisition module 301 is configured to acquire line data from a power distribution network, perform standardized processing on the line data, and generate to-be-detected line data;
a category determining module 302, configured to classify the line data to be detected based on multiple subclasses to obtain a category of the line data to be detected;
a target support vector regression machine selection module 303, configured to select a target support vector regression machine from the multiple support vector regression machines to be selected according to the type of the line data to be detected;
the plurality of subclasses and the plurality of support vector regression machines to be selected are generated through a preset support vector regression machine training module;
and a line loss calculation module 304, configured to input the line data to be detected to the target support vector regression machine, and determine a line loss corresponding to the line data to be detected.
Optionally, the support vector regression training module includes:
the training sample set acquisition submodule is used for acquiring a training sample set;
the standardized sample set generation submodule is used for carrying out standardized processing on the training samples in the training sample set to generate a standardized sample set; the normalized sample set includes a plurality of normalized samples;
the subclass division submodule is used for classifying the standardized sample set by adopting a preset classifier to obtain a plurality of subclasses;
and the support vector regression training submodule is used for training a plurality of preset initial support vector regression machines by adopting standardized samples in a plurality of subclasses to obtain a plurality of support vector regression machines to be selected.
Optionally, the sub-classification sub-module includes:
a first centroid sample selection unit, configured to select a first target normalized sample from the normalized sample set as a first centroid sample using a preset classifier;
a first euclidean distance calculating unit for calculating a first euclidean distance between each of the normalized samples and the first centroid sample;
a classification unit, configured to classify a class of the normalized samples, for which the first euclidean distance is less than or equal to a first preset clustering radius, into a subclass and delete the subclass from the set of normalized samples;
a second centroid sample selection unit, configured to select a second target normalized sample from normalized samples whose first euclidean distance is greater than the first preset clustering radius as a second centroid sample;
a loop unit, configured to update the first centroid sample with the second centroid sample, and return to the step of calculating the first euclidean distance between each normalized sample and the first centroid sample until all normalized samples have a corresponding subclass.
Optionally, the plurality of sub-classes respectively have corresponding class centroid samples, and the class determination module 302 includes:
the second Euclidean distance calculation submodule is used for calculating second Euclidean distances between the line data to be detected and each class centroid sample respectively;
and the class determination submodule is used for taking the subclass corresponding to the class centroid sample of which the second Euclidean distance is smaller than or equal to a second preset clustering radius as the class of the line data to be detected.
An embodiment of the present invention further provides an electronic device, which includes a memory and a processor, where the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method for determining a line loss of a power distribution network according to any one of the above descriptions.
An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by the processor, implements the line loss determination method for the power distribution network according to any one of the above methods.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for determining line loss of a power distribution network is characterized by comprising the following steps:
acquiring line data from a power distribution network, and carrying out standardized processing on the line data to generate line data to be detected;
classifying the line data to be detected based on a plurality of subclasses to obtain the class of the line data to be detected;
selecting a target support vector regression machine from a plurality of support vector regression machines to be selected according to the category of the line data to be detected; the plurality of subclasses and the plurality of support vector regression machines to be selected are generated through a preset support vector regression machine training process;
and inputting the line data to be detected into the target support vector regression machine, and determining the line loss corresponding to the line data to be detected.
2. The method of claim 1, wherein the support vector regression training process comprises:
acquiring a training sample set;
carrying out standardization processing on the training samples in the training sample set to generate a standardized sample set; the normalized sample set includes a plurality of normalized samples;
classifying the standardized sample set by adopting a preset classifier to obtain a plurality of subclasses;
and respectively training a plurality of preset initial support vector regression machines by adopting standardized samples in a plurality of subclasses to obtain a plurality of support vector regression machines to be selected.
3. The method of claim 2, wherein the step of classifying the normalized sample set using a preset classifier to obtain a plurality of subclasses comprises:
selecting a first target normalized sample from the normalized sample set as a first centroid sample using a preset classifier;
calculating a first Euclidean distance between each of the normalized samples and the first centroid sample;
classifying the category of the standardized samples with the first Euclidean distance smaller than or equal to a first preset clustering radius into a subclass and deleting the subclass from the standardized sample set;
selecting a second target normalized sample as a second centroid sample from the normalized samples having the first Euclidean distance greater than the first preset clustering radius;
updating the first centroid sample with the second centroid sample, returning to the step of calculating the first euclidean distance between each of the normalized samples and the first centroid sample until all of the normalized samples have a corresponding subclass.
4. The method according to claim 1, wherein the plurality of subclasses respectively have corresponding class centroid samples, and the step of classifying the line data to be detected based on the plurality of subclasses to obtain the class of the line data to be detected comprises:
respectively calculating a second Euclidean distance between the line data to be detected and each class centroid sample;
and taking the subclass corresponding to the class centroid sample with the second Euclidean distance smaller than or equal to a second preset clustering radius as the class of the line data to be detected.
5. A line loss determination device for a power distribution network, comprising:
the line data acquisition module to be detected is used for acquiring line data from the power distribution network, and carrying out standardized processing on the line data to generate line data to be detected;
the category determining module is used for classifying the line data to be detected based on a plurality of subclasses to obtain the category of the line data to be detected;
the target support vector regression selection module is used for selecting a target support vector regression from a plurality of support vector regression to be selected according to the type of the line data to be detected; the plurality of subclasses and the plurality of support vector regression machines to be selected are generated through a preset support vector regression machine training module;
and the line loss calculation module is used for inputting the line data to be detected into the target support vector regression machine and determining the line loss corresponding to the line data to be detected.
6. The apparatus of claim 5, wherein the support vector regression training module comprises:
the training sample set acquisition submodule is used for acquiring a training sample set;
the standardized sample set generation submodule is used for carrying out standardized processing on the training samples in the training sample set to generate a standardized sample set; the normalized sample set includes a plurality of normalized samples;
the subclass division submodule is used for classifying the standardized sample set by adopting a preset classifier to obtain a plurality of subclasses;
and the support vector regression training submodule is used for training a plurality of preset initial support vector regression machines by adopting standardized samples in a plurality of subclasses to obtain a plurality of support vector regression machines to be selected.
7. The apparatus of claim 6, wherein the sub-partitioning sub-module comprises:
a first centroid sample selection unit, configured to select a first target normalized sample from the normalized sample set as a first centroid sample using a preset classifier;
a first euclidean distance calculating unit for calculating a first euclidean distance between each of the normalized samples and the first centroid sample;
a classification unit, configured to classify a class of the normalized samples, for which the first euclidean distance is less than or equal to a first preset clustering radius, into a subclass and delete the subclass from the set of normalized samples;
a second centroid sample selection unit, configured to select a second target normalized sample from normalized samples whose first euclidean distance is greater than the first preset clustering radius as a second centroid sample;
a loop unit, configured to update the first centroid sample with the second centroid sample, and return to the step of calculating the first euclidean distance between each normalized sample and the first centroid sample until all normalized samples have a corresponding subclass.
8. The apparatus of claim 5, wherein the plurality of sub-classes each have a corresponding class centroid sample, and wherein the class determination module comprises:
the second Euclidean distance calculation submodule is used for calculating second Euclidean distances between the line data to be detected and each class centroid sample respectively;
and the class determination submodule is used for taking the subclass corresponding to the class centroid sample of which the second Euclidean distance is smaller than or equal to a second preset clustering radius as the class of the line data to be detected.
9. An electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and wherein the computer program, when executed by the processor, causes the processor to perform the steps of the method for determining line loss for an electrical distribution network according to any one of claims 1-4.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out a method for determining a line loss of an electrical distribution network according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011531544.2A CN112632857A (en) | 2020-12-22 | 2020-12-22 | Method, device, equipment and storage medium for determining line loss of power distribution network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011531544.2A CN112632857A (en) | 2020-12-22 | 2020-12-22 | Method, device, equipment and storage medium for determining line loss of power distribution network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112632857A true CN112632857A (en) | 2021-04-09 |
Family
ID=75321090
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011531544.2A Pending CN112632857A (en) | 2020-12-22 | 2020-12-22 | Method, device, equipment and storage medium for determining line loss of power distribution network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112632857A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113325231A (en) * | 2021-05-21 | 2021-08-31 | 国网山东省电力公司济南供电公司 | Platform district line loss monitoring analysis positioner |
| CN115616333A (en) * | 2022-12-20 | 2023-01-17 | 国网江西省电力有限公司电力科学研究院 | Power distribution network line loss prediction method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2014027737A (en) * | 2012-07-25 | 2014-02-06 | Hokuriku Electric Power Co Inc:The | Method and device for calculating setting value of line voltage drop compensator |
| CN106780121A (en) * | 2016-12-06 | 2017-05-31 | 广州供电局有限公司 | A kind of multiplexing electric abnormality recognition methods based on power load pattern analysis |
| CN107167702A (en) * | 2017-05-04 | 2017-09-15 | 国网福建省电力有限公司 | A kind of distribution feeder fault type recognition method and device |
| US20180322416A1 (en) * | 2016-08-30 | 2018-11-08 | Soochow University | Feature extraction and classification method based on support vector data description and system thereof |
| CN110110887A (en) * | 2019-03-22 | 2019-08-09 | 国网浙江省电力有限公司信息通信分公司 | To the prediction technique of low-voltage platform area line loss per unit |
| CN111062608A (en) * | 2019-12-14 | 2020-04-24 | 贵州电网有限责任公司 | Line loss monitoring method for 10kV line based on line loss classifier |
-
2020
- 2020-12-22 CN CN202011531544.2A patent/CN112632857A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2014027737A (en) * | 2012-07-25 | 2014-02-06 | Hokuriku Electric Power Co Inc:The | Method and device for calculating setting value of line voltage drop compensator |
| US20180322416A1 (en) * | 2016-08-30 | 2018-11-08 | Soochow University | Feature extraction and classification method based on support vector data description and system thereof |
| CN106780121A (en) * | 2016-12-06 | 2017-05-31 | 广州供电局有限公司 | A kind of multiplexing electric abnormality recognition methods based on power load pattern analysis |
| CN107167702A (en) * | 2017-05-04 | 2017-09-15 | 国网福建省电力有限公司 | A kind of distribution feeder fault type recognition method and device |
| CN110110887A (en) * | 2019-03-22 | 2019-08-09 | 国网浙江省电力有限公司信息通信分公司 | To the prediction technique of low-voltage platform area line loss per unit |
| CN111062608A (en) * | 2019-12-14 | 2020-04-24 | 贵州电网有限责任公司 | Line loss monitoring method for 10kV line based on line loss classifier |
Non-Patent Citations (5)
| Title |
|---|
| 丁忠安等: "基于专家样本库和最小二乘支持向量机的配电网线损率预测模型", 《水电能源科学》 * |
| 史琳等: "基于多源数据互补技术的线损智能分析平台研究", 《自动化与仪表》 * |
| 徐晖等: "基于机器学习的配电网线损自动计算模型", 《电力行业信息化优秀论文集2013》 * |
| 王恩等: "基于谱质心迁移的线损优化分摊方法研究", 《智慧电力》 * |
| 黄训诚等: "基于支持向量回归的配电网线损计算模型", 《电力系统保护与控制》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113325231A (en) * | 2021-05-21 | 2021-08-31 | 国网山东省电力公司济南供电公司 | Platform district line loss monitoring analysis positioner |
| CN113325231B (en) * | 2021-05-21 | 2022-06-24 | 国网山东省电力公司济南供电公司 | Transformer area line loss monitoring, analyzing and positioning device |
| CN115616333A (en) * | 2022-12-20 | 2023-01-17 | 国网江西省电力有限公司电力科学研究院 | Power distribution network line loss prediction method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111505433B (en) | Low-voltage transformer area indoor variable relation error correction and phase identification method | |
| Keselman et al. | Many-to-many graph matching via metric embedding | |
| CN111612275B (en) | Method and device for predicting load of regional user | |
| CN112632857A (en) | Method, device, equipment and storage medium for determining line loss of power distribution network | |
| CN107633257B (en) | Data quality evaluation method and device, computer readable storage medium and terminal | |
| CN112819649B (en) | Method and device for determining user change relation of station area | |
| CN111537884A (en) | Method and device for acquiring service life data of power battery, computer equipment and medium | |
| CN108770002A (en) | Base station flow analysis method, device, equipment and storage medium | |
| CN110796159A (en) | Power data classification method and system based on k-means algorithm | |
| US20070198252A1 (en) | Optimum design management apparatus, optimum design calculation system, optimum design management method, and optimum design management program | |
| CN112765362A (en) | Knowledge graph entity alignment method based on improved self-encoder and related equipment | |
| CN108898273A (en) | A kind of user side load characteristic clustering evaluation method based on morphological analysis | |
| CN106250918A (en) | A kind of mixed Gauss model matching process based on the soil-shifting distance improved | |
| CN104778205B (en) | A kind of mobile application sequence and clustering method based on Heterogeneous Information network | |
| CN106651630B (en) | Key electricity utilization industry identification method and system | |
| CN111798152A (en) | Intelligent store management method and device | |
| CN116363416A (en) | Image de-duplication method and device, electronic equipment and storage medium | |
| CN116662840A (en) | Low-voltage station user phase identification method based on machine learning | |
| CN115661472A (en) | Image duplicate checking method and device, computer equipment and storage medium | |
| CN109919811B (en) | Insurance agent culture scheme generation method based on big data and related equipment | |
| CN113487080A (en) | Wind speed dynamic scene generation method, system and terminal based on wind speed classification | |
| CN110852767A (en) | Passenger flow volume clustering method and terminal equipment | |
| Thakur et al. | Object Recognition From Image Using Grid Based Color Moments Feature Extraction Method | |
| CN116049700B (en) | Multi-mode-based operation and inspection team portrait generation method and device | |
| CN114462901B (en) | Active power distribution network electric energy quality evaluation method and device, terminal and storage medium |
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: 20210409 |
|
| RJ01 | Rejection of invention patent application after publication |