CN113095692A - Comprehensive evaluation method and system for operation quality of power distribution network circuit considering user requirements - Google Patents
Comprehensive evaluation method and system for operation quality of power distribution network circuit considering user requirements Download PDFInfo
- Publication number
- CN113095692A CN113095692A CN202110416967.8A CN202110416967A CN113095692A CN 113095692 A CN113095692 A CN 113095692A CN 202110416967 A CN202110416967 A CN 202110416967A CN 113095692 A CN113095692 A CN 113095692A
- Authority
- CN
- China
- Prior art keywords
- score
- index
- line
- evaluation
- point
- 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
- 238000011156 evaluation Methods 0.000 title claims abstract description 94
- 239000013598 vector Substances 0.000 claims abstract description 51
- 238000013528 artificial neural network Methods 0.000 claims abstract description 19
- 238000012423 maintenance Methods 0.000 claims abstract description 13
- 238000012847 principal component analysis method Methods 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims description 40
- 238000000034 method Methods 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 14
- 230000007547 defect Effects 0.000 claims description 13
- 230000002159 abnormal effect Effects 0.000 claims description 12
- 230000009467 reduction Effects 0.000 claims description 9
- 230000009471 action Effects 0.000 claims description 6
- 230000011664 signaling Effects 0.000 claims description 6
- 230000007781 signaling event Effects 0.000 claims description 6
- 150000001875 compounds Chemical class 0.000 claims description 3
- 239000004020 conductor Substances 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000012935 Averaging Methods 0.000 claims description 2
- 238000013441 quality evaluation Methods 0.000 abstract description 8
- 230000009466 transformation Effects 0.000 abstract 2
- 238000004422 calculation algorithm Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 239000002131 composite material Substances 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000013210 evaluation model Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Entrepreneurship & Innovation (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Bioinformatics & Computational Biology (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a comprehensive evaluation method and a system for the operation quality of a power distribution network line considering user requirements, which provide a more scientific operation quality evaluation result for reasonable maintenance, and comprises the steps of constructing an evaluation index vector, wherein the evaluation index vector comprises a distribution transformation automation index, a commissioning age index, a user satisfaction rate index, a distribution transformation power quality index, a reliability index and a heavy overload index, grading each index in the evaluation index vector, converting the evaluation index vector into a principal component vector by adopting a principal component analysis method, calculating the contribution rate of each principal component in the principal component vector according to a characteristic value, and calculating the weight of a corresponding evaluation index according to the contribution rate of each principal component; calculating a comprehensive score according to the score and the weight of each evaluation index; and grading the operation quality of the power distribution network line according to the comprehensive grade. An evaluation system is constructed based on a neural network and by using historical evaluation results, the end-to-end connection of the original indexes and the final scores is realized, the operation rate is optimized, and the cost is reduced.
Description
Technical Field
The invention relates to the technical field of power distribution network operation, in particular to an evaluation method for the operation quality of a power distribution network line.
Background
The distribution line is used as a network system for connecting a power supply system and a power consumer, and is an important link for ensuring the stable operation of a power grid. Because of direct connection with the user, the distribution line has characteristics such as wide distribution range, system complicacy and operation maintenance difficulty. The operation quality of the distribution line is affected by factors such as large quantity, poor operation environment and the like, and the problems that the distribution line has high fault probability, poor power supply reliability and the like are caused. And the distribution line maintenance can effectively reduce the probability of fault occurrence, and has important significance for maintaining the stable operation of the power grid. Aiming at the maintenance of the 10kV distribution line, in order to avoid the problems of excessive maintenance and lack of maintenance, the reasonable evaluation of the line quality is an important guarantee for correct and effective decision. Aiming at the rapid development of the distribution lines, in order to ensure the power utilization quality and meet the requirements of users, the traditional quality evaluation method is gradually difficult to meet the actual requirements, and a more perfect evaluation system is needed to ensure the reliability of the distribution lines.
Disclosure of Invention
Aiming at the technical defects, the invention provides a comprehensive evaluation method for the operation quality of a power distribution network line considering user requirements, and solves the technical problem of how to provide a more scientific operation quality evaluation result for reasonable maintenance.
In order to solve the technical problems, the technical scheme of the invention is as follows: a comprehensive evaluation method for the operation quality of a power distribution network circuit considering user requirements comprises the following steps:
constructing an evaluation index vector comprising equipment indexes and effect indexes, wherein the dimension of the evaluation index vector is n, the equipment indexes comprise distribution automation indexes and commissioning age indexes, and the effect indexes comprise user satisfaction rate indexes, distribution power quality indexes, reliability indexes and heavy overload indexes;
scoring each index in the evaluation index vector according to a scoring standard, and constructing a sample data matrix according to the scoring vector of the evaluation index vector for n years, wherein an element x in the sample data matrixijA score representing the jth index for the ith year;
extracting an eigenvalue and an eigenvector matrix of a sample data matrix, converting an evaluation index vector into a principal component vector by adopting a principal component analysis method, wherein each principal component in the principal component vector corresponds to an evaluation index in the evaluation index vector one by one, then calculating the contribution rate of each principal component in the principal component vector according to the eigenvalue, and respectively calculating the weight of the corresponding evaluation index according to the contribution rate of each principal component and the eigenvector matrix;
calculating a comprehensive score according to the score and the weight of each evaluation index in the evaluation index vector;
and grading the operation quality of the power distribution network line according to the grading interval in which the comprehensive grade falls.
Further, the feature vector matrix is constructed as follows:
carrying out standardization processing on a sample data matrix:
in the formula (I), the compound is shown in the specification,
represents the element xijA normalized value of (d);
solving the correlation coefficient rij：
Constructing a correlation coefficient matrix R according to the correlation coefficients, wherein the correlation coefficients RijIs the ith row and the jth column element in the correlation coefficient matrix R; calculating characteristic value (lambda) of correlation coefficient matrix R by adopting Jacobian method1，λ2，...，λn) And corresponding eigenvector, and further obtaining an eigenvector matrix A, aijRepresenting the ith row and jth column element in the eigenvector matrix a.
Further, the weight calculation formula of the evaluation index is as follows:
in the formula, wiA weight indicating an ith evaluation index; beta is aiIndicating the contribution ratio of the ith principal component.
Further, the calculation formula of the composite score is as follows:
in the formula, wiA weight indicating an ith evaluation index; x is the number ofiIndicates the ith evaluation index XiThe score of (1).
Further, the scoring criteria were as follows:
the scoring criteria were as follows:
X1representing the line operation index, the total score is 55 scores, is synthesized by 10 indexes, and is respectively:
(1) n-1 index, the line is transferred to more than 3 lines, and the mark is 10 points; 2, 8 points are marked; 1, marking 6 points; if not, recording 1 point;
(2) the high current-carrying capacity conductor section accounts for a ratio, the full score is 5 minutes, and the calculation rule is as follows: 5 × length of overhead line (km) below 170mm 2/total length of overhead line (km) × 100% +1-185mm 2/total length of cable line (km) × 100%));
(3) the abnormal operation mode of the line has no abnormal mode, the full score is 5, the abnormal mode exists, and 0.5 is deducted from 1 abnormal operation mode. If the final score is less than 1 point, recording 1 point;
(4) the circuit has running defects, no running defects are generated, and the full score is 5 points; if the operation defect exists, 1 operation defect is deducted by 1 point, and if the final point is less than 1 point, the point is recorded by 1 point;
(5) under the condition of cables in the same channel, the maximum number of isomorphic cables related to a channel of a line is less than or equal to 5, and the full score is 5; 6-10, 4 points are marked; 11-13, 3 points are marked; 13-16, 2 points, 16 or more, 1 point.
(6) Under the condition of line heavy overload (5 minutes), the line no-heavy overload is counted for 5 minutes, and the main line heavy overload is buckled for 0.5 minute once; if the final score is less than 1 point, recording 1 point;
(7) under the condition of low voltage of the line, the full score is 5, and the calculation rule is as follows: (1-subscriber of low voltage line/subscriber of line) 5
(8) The commissioning time is 5 minutes to full within 5 years according to the longest equipment of the line running time; and (4) deducting theta every 1 year after more than 5 years, and recording 1 point if the final point is less than 1 point.
(9) And (4) protection configuration, wherein the circuit has three sections of protection, the full score is 5, the two sections are 4, the first section is 3, and the no section is 0.
(10) And (4) planning maintenance times, wherein within 12 months, the planned maintenance is increased by 1 point every time, and the maximum time is 5 points.
X2The customer satisfaction rate is represented as 10 points, the number of complains per each occurrence in 12 months is 200/line user (household), and the number of complains per each occurrence in 12 to 24 months is 100/line user (household). If the final score is less than 1 point, then the score is 1.
X3Representing the reliability of power supply, wherein within 12 months, the number of households is 0 when power is off, and the full score is 15; the number of households is reduced by 80 to 99 percent compared with the number of households in the last year in power failure, and the number of households is 13 minutes; the reduction is 60 to 79 percent compared with the last year, and 11 points are obtained; the reduction is 40-59% in the last year, and 10 points are obtained; the year is 20-39% less than the last year, and 9 points are obtained; the reduction is 0 to 19 percent compared with the last year, and 8 points are obtained; the increase is 1-20% in the last year, and the score is 6; the increase is 21-40% in the last year, and 5 points are obtained; the increase is 41-60% in the last year, and 3 points are obtained; the increase is 61% or more in the last year, and the score is 1.
X4Representative of power distributionThe method comprises the following steps of (1) dynamically operating the system, fully dividing the system into 15 points, dividing the indexes into a main station operation index and a terminal operation index, wherein in the main station operation index, a remote signaling action score is equal to a remote signaling action accuracy rate of 2, a remote control action score is equal to a remote control action accuracy rate of 2, and an FA success score is equal to an FA success rate of 2; in the terminal operation indexes, the terminal on-line score is terminal on-line rate 2, the remote control execution score is remote control execution accuracy 3, the remote signaling collection score is remote signaling collection accuracy 2, and the terminal fault rate score (2 points) is terminal fault rate 2;
X5representing a line fault index, and deducting 1 point when a main line trips within 12 months; the branch switch tripping button is 0.5 minutes; and 0.1 minute is deducted when the line belongs to the transformer area and has a primary fault. If the final score is less than 1 point, then the score is 1. The invention also provides a comprehensive evaluation system for the operation quality of the power distribution network line considering the user requirements, which comprises a rating neural network; the comprehensive evaluation method for the operation quality of the power distribution network line, which is provided by the invention and takes the user requirements into consideration, is used for obtaining the rating result of the operation quality of the power distribution network line, and constructing a training sample set, wherein the training samples in the training sample set take the evaluation index vector as input, and take the corresponding rating result as a label;
and training the rating neural network through the training sample set, so that the rating neural network has the capability of inputting an evaluation index vector and outputting a rating result.
Furthermore, the rating neural network comprises a plurality of weak classifiers and a strong classifier, wherein the weak classifiers take the evaluation index vectors as input and output weak classification results to the strong classifier; and the strong classifier outputs a final rating result after weighting and averaging the weak classification result.
Furthermore, a BP neural network is adopted as a weak classifier, and an Adaboost strong classifier is adopted.
Compared with the prior art, the invention has the advantages that:
1. the evaluation indexes selected by the method are wider in coverage range and larger in information quantity, important indexes such as power supply capacity, power supply quality and power supply reliability can be fully extracted, user requirements are fully respected (whether the user requirements are met or not is reflected through user satisfaction rate), meanwhile objective indexes are combined to comprehensively evaluate the operation quality of the power distribution network, more scientific operation quality evaluation is realized, and excessive overhaul or omission of overhaul caused by unreasonable evaluation is avoided. Therefore, the evaluation system constructed by the invention has incomparable advancement compared with the prior art, has the characteristics of exquisite structure, high adaptability and strong comprehensiveness, and can realize the comprehensive evaluation of the running quality of the line.
2. The comprehensive scoring is carried out according to each evaluation index, and the influence degree of each evaluation index on the operation quality is determined by a principal component analysis method, so that the comprehensive scoring is more scientific and reasonable. And then grading according to the grades to accord with the actual operation condition of the distribution lines, wherein the distribution operation quality distribution of each region is concentrated, and the grades of the distribution networks in the same region are slightly different but fall into the same grade.
3. The invention also provides an evaluation system based on the neural network, the neural network is trained by using the historical evaluation result, the obtained line quality evaluation model greatly simplifies the complex calculation process of the traditional method, and the original index and the final grade are connected end to end by adopting a deep learning method. The model can optimize the operation rate and reduce the cost while ensuring the evaluation precision.
Drawings
Fig. 1 is a flow chart of distribution line index weight determination.
Fig. 2 is a flow chart of the establishment of the distribution line planning quality evaluation system.
Detailed Description
One) construction of evaluation index vector
In the present embodiment, a 10kV distribution network is taken as an example, and 5 evaluation indexes are selected, so that the dimension n of the evaluation index vector is 5, and the evaluation indexes are respectively scored. The rationality of the indexes directly influences the quality of the result, and the distribution line planning quality level can be comprehensively reflected. The evaluation index vector comprises a line operation index, a user satisfaction rate index, power supply reliability, a power distribution automation operation level and a line fault index, and the scoring standards of the indexes are as follows:
X1representing the line operation index, the total score is 55 scores, is synthesized by 10 indexes, and is respectively:
(1) n-1 index, the line is transferred to more than 3 lines, and the mark is 10 points; 2, 8 points are marked; 1, marking 6 points; if not, recording 1 point;
(2) the high current-carrying capacity conductor section accounts for a ratio, the full score is 5 minutes, and the calculation rule is as follows: 5 x (length of overhead line (kilometer) below 1-70mm 2/total length of overhead line (kilometer) × 100% + length of cable line (kilometer) below 1-185mm 2/total length of cable line (kilometer) × 100%));
(3) the abnormal operation mode of the line has no abnormal mode, the full score is 5, the abnormal mode exists, and 0.5 is deducted from 1 abnormal operation mode. If the final score is less than 1 point, recording 1 point;
(4) the circuit has running defects, no running defects are generated, and the full score is 5 points; if the operation defect exists, 1 operation defect is deducted by 1 point, and if the final point is less than 1 point, the point is recorded by 1 point;
(5) under the condition of cables in the same channel, the maximum number of isomorphic cables related to a channel of a line is less than or equal to 5, and the full score is 5; 6-10, 4 points are marked; 11-13, 3 points are marked; 13-16, 2 points, 16 or more, 1 point.
(6) Under the condition of line heavy overload (5 minutes), the line no-heavy overload is counted for 5 minutes, and the main line heavy overload is buckled for 0.5 minute once; if the final score is less than 1 point, recording 1 point;
(7) under the condition of low voltage of the line, the full score is 5, and the calculation rule is as follows: (1-subscriber of low voltage line/subscriber of line) 5
(8) The commissioning time is 5 minutes to full within 5 years according to the longest equipment of the line running time; and (4) deducting 0.2 point every 1 year after more than 5 years, and if the final point is less than 1 point, recording 1 point.
(9) And (4) protection configuration, wherein the circuit has three sections of protection, the full score is 5, the two sections are 4, the first section is 3, and the no section is 0.
(10) And (4) planning maintenance times, wherein within 12 months, the planned maintenance is increased by 1 point every time, and the maximum time is 5 points.
X2The satisfaction rate of the user is represented, and the non-occurrence of complaints is 10 minutes of full score and 12 monthsThe deduction of each complaint in 12 to 24 months is 200/line user number (household). If the final score is less than 1 point, then the score is 1.
X3Representing the reliability of power supply, wherein within 12 months, the number of households is 0 when power is off, and the full score is 15; the number of households is reduced by 80 to 99 percent compared with the number of households in the last year in power failure, and the number of households is 13 minutes; the reduction is 60 to 79 percent compared with the last year, and 11 points are obtained; the reduction is 40-59% in the last year, and 10 points are obtained; the year is 20-39% less than the last year, and 9 points are obtained; the reduction is 0 to 19 percent compared with the last year, and 8 points are obtained; the increase is 1-20% in the last year, and the score is 6; the increase is 21-40% in the last year, and 5 points are obtained; the increase is 41-60% in the last year, and 3 points are obtained; the increase is 61% or more in the last year, and the score is 1.
X4Representing the distribution automation operation level, fully dividing the index into a main station operation index and a terminal operation index by 15 points, wherein in the main station operation index, the remote signaling action score is equal to the remote signaling action accuracy rate of 2, the remote control action score is equal to the remote control action accuracy rate of 2, and the FA success score is equal to the FA success rate of 2; in the terminal operation indexes, the terminal on-line score is terminal on-line rate 2, the remote control execution score is remote control execution accuracy 3, the remote signaling collection score is remote signaling collection accuracy 2, and the terminal fault rate score (2 points) is terminal fault rate 2;
X5representing a line fault index, and deducting 1 point when a main line trips within 12 months; the branch switch tripping button is 0.5 minutes; and 0.1 minute is deducted when the line belongs to the transformer area and has a primary fault. If the final score is less than 1 point, then the score is 1.
Scoring each index in the evaluation index vector according to a scoring standard, and constructing a sample data matrix according to the scoring vector of the evaluation index vector for 6 years, wherein an element x in the sample data matrixijA score representing the jth index for the ith year; the sample data matrix X is as follows:
wherein x isijThe score obtained for the jth index in year i (i ═ 1, 2,...，5；j＝1，2，...，5)。
Whether the setting of the index weight is reasonable or not directly influences the final evaluation result. The invention adopts a principal component analysis method to analyze the influence degree of each index. In order to improve the reliability level of the weight result, it is necessary to determine and verify whether the setting of the index weight is reasonable under certain objective condition constraints. The principal component analysis method is a statistical method which recombines the original indexes into a group of new independent comprehensive indexes to replace the original indexes. In view of the above, the method for evaluating the line quality provided by the invention utilizes a principal component analysis method to construct an evaluation model of the line quality, and objective statistical analysis is carried out on the evaluation index of the 10kV line operation quality, so that the evaluation method is more objective.
Two) calculating the index weight
Referring to fig. 1, the index weight calculation process is as follows: extracting the eigenvalue and the eigenvector matrix of the sample data matrix, converting the evaluation index vector into principal component vectors by adopting a principal component analysis method, wherein each principal component in the principal component vectors corresponds to the evaluation index in the evaluation index vector one by one, then calculating the contribution rate of each principal component in the principal component vectors according to the eigenvalue, and respectively calculating the weight of the corresponding evaluation index according to the contribution rate of each principal component and the eigenvector matrix.
The eigenvector matrix is constructed as follows:
carrying out standardization processing on a sample data matrix:
in the formula (I), the compound is shown in the specification,
represents the element xijA normalized value of (d);
solving the correlation coefficient rij：
Constructing a correlation coefficient matrix R according to the correlation coefficients, wherein the correlation coefficients RijIs the ith row and the jth column element in the correlation coefficient matrix R; calculating characteristic value (lambda) of correlation coefficient matrix R by adopting Jacobian method1，λ2，...，λn) And corresponding eigenvector, and further obtaining an eigenvector matrix A, aijRepresenting the ith row and jth column element in the eigenvector matrix a.
Obtaining a sample correlation coefficient matrix:
solving characteristic value (lambda) of the decorrelation coefficient R by adopting the Jacobian method1，λ2，...，λ6) And corresponding feature vector aij(i 1, 2.., 5; j 1, 2.., 5), further deriving a feature vector matrix:
converting 5 variables of 5 samples into original variable Xi(i ═ 1, 2.., 5) into a principal component Fi(i ═ 1, 2.., 5), the principal component is a linear combination of the original variables, i.e.:
solving the contribution rate of each principal component:
the contribution rate is the proportion of the variance of a certain principal component to all the variances, and actually is the proportion of a certain eigenvalue to all the eigenvalues, namely:
the weights of the indexes are calculated as follows:
three) comprehensive evaluation
Line operation quality rating standard
After the scoring weight of each index is calculated, the comprehensive score can be calculated:
wherein X is the composite score, X is the same score of 10 pointsiA score for each i index.
According to the actual operation condition of the distribution line, the distribution operation quality distribution of each area is concentrated, so that a new distribution line quality index rating standard is determined as shown in the following table:
table 1 distribution line operation quality index rating standard
| Scoring | 10-9 | 9-8 | 8-7 | 7-6 | 6-5.5 | 5.5-0 |
| Rating | A+ | A | B | C | D | E |
The method comprises the steps of grading A + and A, wherein the grades A + and A are power distribution network planning quality high-quality demonstration areas, the grades B are power distribution network planning quality well-developed areas, the grades C are power distribution network planning quality typical horizontal areas, the grades D are power distribution network planning quality honest areas to be improved, and the grades E are power distribution network planning quality key monitoring areas.
Four) comprehensive quality evaluation system is provided
Referring to fig. 2, the operation flow is simplified and the performance of the evaluation system is checked. The invention adopts an Adaboost strong classifier based on a BP neural network to construct a comprehensive evaluation system for the operation quality of the power distribution network circuit considering the user requirements. The idea of the Adaboost algorithm is to combine the outputs of multiple "weak" classifiers to produce an efficient classification. The BP-Adaboost model takes a BP neural network as a weak classifier, repeatedly trains the BP neural network to predict sample output, and obtains a strong classifier consisting of a plurality of BP neural network weak classifiers through an Adaboost algorithm. The method mainly comprises the following steps:
1) data selection and network initialization: firstly, a weak learning algorithm and a sample space (x, y) are given, m groups of training data are found from the sample space, and the weight of each group of training data is 1/m.
2) And (3) weak classifier prediction, calculating a predicted sequence weight: and (3) computing for T times by adopting a weak learning algorithm iteration mode, namely updating the weight distribution of the training data according to the classification result after each computation, giving a larger weight to the training individuals with failed classification, and paying more attention to the training individuals during the next iteration computation.
3) Adjusting the test weight: the weak classifier obtains a classification function sequence f1, f2... fT through repeated iteration, each classification function is endowed with a weight, and the better the classification result is, the larger the corresponding weight is. After T iterations, the final strong classification function F is weighted by the weak classification function.
In the experiment, six original indexes are used as samples of the neural network to be input, and six grades divided by the obtained total grades are used as labels to train the BP-Adaboost neural network. The obtained line quality evaluation model greatly simplifies the complex calculation process of the traditional method, and the deep learning method is adopted to realize the end-to-end connection of the original index and the final score. The model can optimize the operation rate and reduce the cost while ensuring the evaluation precision.
Claims (10)
1. A comprehensive evaluation method for the operation quality of a power distribution network circuit considering user requirements is characterized by comprising the following steps:
constructing an evaluation index vector, wherein the dimension of the evaluation index vector is n, and the evaluation index vector comprises a line operation index, a user satisfaction rate index, power supply reliability, a power distribution automation operation level and a line fault index;
scoring each index in the evaluation index vector according to a scoring standard, and constructing a sample data matrix according to the scoring vector of the evaluation index vector for n years, wherein an element x in the sample data matrixijA score representing the jth index for the ith year;
extracting an eigenvalue and an eigenvector matrix of a sample data matrix, converting an evaluation index vector into a principal component vector by adopting a principal component analysis method, wherein each principal component in the principal component vector corresponds to an evaluation index in the evaluation index vector one by one, then calculating the contribution rate of each principal component in the principal component vector according to the eigenvalue, and respectively calculating the weight of the corresponding evaluation index according to the contribution rate of each principal component and the eigenvector matrix;
calculating a comprehensive score according to the score and the weight of each evaluation index in the evaluation index vector;
and grading the operation quality of the power distribution network line according to the grading interval in which the comprehensive grade falls.
2. The comprehensive evaluation method for the operation quality of the power distribution network circuit considering the user demand according to claim 1, characterized in that the eigenvector matrix is constructed as follows:
carrying out standardization processing on a sample data matrix:
in the formula (I), the compound is shown in the specification,
represents the element xijA normalized value of (d);
solving the correlation coefficient rij：
Constructing a correlation coefficient matrix R according to the correlation coefficients, wherein the correlation coefficients RijIs the ith row and the jth column element in the correlation coefficient matrix R; calculating characteristic value (lambda) of correlation coefficient matrix R by adopting Jacobian method1，λ2，...，λn) And corresponding eigenvector, and further obtaining an eigenvector matrix A, aijRepresenting the ith row and jth column element in the eigenvector matrix a.
3. The comprehensive evaluation method for the operation quality of the power distribution network circuit considering the user demand according to claim 2, wherein the general formula of the main components is as follows:
Fi＝ai1X1+ai2X2+...+aijXj+...+ainXn；
in the formula, FiThe ith evaluation index X in the expression and evaluation index vectoriA corresponding principal component; xjVector representing evaluation indexThe j-th evaluation index; 1, 2, n; j is 1, 2.
4. The comprehensive evaluation method for the operation quality of the power distribution network circuit considering the user demand according to claim 2, characterized in that the calculation formula of the contribution rate is as follows:
in the formula, λiAn ith eigenvalue representing a correlation coefficient matrix R; beta is aiIndicating the contribution ratio of the ith principal component.
5. The comprehensive evaluation method for the operation quality of the power distribution network circuit considering the user requirements as claimed in claim 1, wherein the weight calculation formula of the evaluation index is as follows:
in the formula, wiA weight indicating an ith evaluation index; beta is aiIndicating the contribution ratio of the ith principal component.
6. The method for comprehensively evaluating the operation quality of the power distribution network circuit considering the user demand according to claim 1, wherein the calculation formula of the comprehensive score is as follows:
in the formula, wiA weight indicating an ith evaluation index; index XiIndicates the ith evaluation index.
7. The comprehensive evaluation method for the operation quality of the power distribution network circuit considering the user demands as claimed in claim 2, wherein the scoring criteria are as follows:
X1representing the line operation index, the total score is 55 scores, is synthesized by 10 indexes, and is respectively:
(1) n-1 index, the line is transferred to more than 3 lines, and the mark is 10 points; 2, 8 points are marked; 1, marking 6 points; if not, recording 1 point;
(2) the high current-carrying capacity conductor section accounts for a ratio, the full score is 5 minutes, and the calculation rule is as follows: 5 x (length of overhead line (kilometer) below 1-70mm 2/total length of overhead line (kilometer) × 100% + length of cable line (kilometer) below 1-185mm 2/total length of cable line (kilometer) × 100%));
(3) the abnormal operation mode of the line has no abnormal mode, the full score is 5, the abnormal mode exists, and 0.5 is deducted from 1 abnormal operation mode. If the final score is less than 1 point, recording 1 point;
(4) the circuit has running defects, no running defects are generated, and the full score is 5 points; if the operation defect exists, 1 operation defect is deducted by 1 point, and if the final point is less than 1 point, the point is recorded by 1 point;
(5) under the condition of cables in the same channel, the maximum number of isomorphic cables related to a channel of a line is less than or equal to 5, and the full score is 5; 6-10, 4 points are marked; 11-13, 3 points are marked; 13-16, 2 points, 16 or more, 1 point.
(6) Under the condition of line heavy overload (5 minutes), the line no-heavy overload is counted for 5 minutes, and the main line heavy overload is buckled for 0.5 minute once; if the final score is less than 1 point, recording 1 point;
(7) under the condition of low voltage of the line, the full score is 5, and the calculation rule is as follows: (1-subscriber of low voltage line/subscriber of line) 5
(8) The commissioning time is 5 minutes to full within 5 years according to the longest equipment of the line running time; and (4) deducting 0.2 point every 1 year after more than 5 years, and if the final point is less than 1 point, recording 1 point.
(9) And (4) protection configuration, wherein the circuit has three sections of protection, the full score is 5, the two sections are 4, the first section is 3, and the no section is 0.
(10) And (4) planning maintenance times, wherein within 12 months, the planned maintenance is increased by 1 point every time, and the maximum time is 5 points.
X2Representing the customer satisfaction rate, no complaints occurredThe full score is 10 points, each complaining deduction generated in 12 months is 200/line user number (household), and each complaining deduction generated in 12 to 24 months is 100/line user number (household). If the final score is less than 1 point, then the score is 1.
X3Representing the reliability of power supply, wherein within 12 months, the number of households is 0 when power is off, and the full score is 15; the number of households is reduced by 80 to 99 percent compared with the number of households in the last year in power failure, and the number of households is 13 minutes; the reduction is 60 to 79 percent compared with the last year, and 11 points are obtained; the reduction is 40-59% in the last year, and 10 points are obtained; the year is 20-39% less than the last year, and 9 points are obtained; the reduction is 0 to 19 percent compared with the last year, and 8 points are obtained; the increase is 1-20% in the last year, and the score is 6; the increase is 21-40% in the last year, and 5 points are obtained; the increase is 41-60% in the last year, and 3 points are obtained; the increase is 61% or more in the last year, and the score is 1.
X4Representing the distribution automation operation level, fully dividing the index into a main station operation index and a terminal operation index by 15 points, wherein in the main station operation index, the remote signaling action score is equal to the remote signaling action accuracy rate of 2, the remote control action score is equal to the remote control action accuracy rate of 2, and the FA success score is equal to the FA success rate of 2; in the terminal operation indexes, the terminal on-line score is terminal on-line rate 2, the remote control execution score is remote control execution accuracy 3, the remote signaling collection score is remote signaling collection accuracy 2, and the terminal fault rate score (2 points) is terminal fault rate 2;
X5representing a line fault index, and deducting 1 point when a main line trips within 12 months; the branch switch tripping button is 0.5 minutes; and 0.1 minute is deducted when the line belongs to the transformer area and has a primary fault. If the final score is less than 1 point, then the score is 1.
8. The comprehensive evaluation method for the operation quality of the power distribution network circuit considering the user requirements as claimed in claim 7, wherein the expression of the principal components is as follows:
in the formula, F1、F2、F3、F4、F5Respectively represent X1、X2、X3、X4、X5The corresponding principal component.
9. A comprehensive evaluation system for the operation quality of a power distribution network line considering user requirements is characterized by comprising a rating neural network; obtaining a rating result of the operation quality of the power distribution network line by the comprehensive evaluation method of the operation quality of the power distribution network line considering the user requirements, according to any claim from 1 to 7, and constructing a training sample set, wherein training samples in the training sample set take evaluation index vectors as input, and take corresponding rating results as labels;
and training the rating neural network through the training sample set, so that the rating neural network has the capability of inputting an evaluation index vector and outputting a rating result.
10. The comprehensive evaluation system for the operation quality of the power distribution network circuit considering the user demand according to claim 9, wherein the rating neural network comprises a plurality of weak classifiers and a strong classifier, the weak classifiers take the evaluation index vector as input and output the weak classification result to the strong classifier; the strong classifier outputs a final rating result after weighting and averaging the weak classification results; a BP neural network is adopted as a weak classifier, and an Adaboost strong classifier is adopted.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110416967.8A CN113095692A (en) | 2021-04-16 | 2021-04-16 | Comprehensive evaluation method and system for operation quality of power distribution network circuit considering user requirements |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110416967.8A CN113095692A (en) | 2021-04-16 | 2021-04-16 | Comprehensive evaluation method and system for operation quality of power distribution network circuit considering user requirements |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113095692A true CN113095692A (en) | 2021-07-09 |
Family
ID=76678524
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110416967.8A Pending CN113095692A (en) | 2021-04-16 | 2021-04-16 | Comprehensive evaluation method and system for operation quality of power distribution network circuit considering user requirements |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113095692A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106991531A (en) * | 2017-03-30 | 2017-07-28 | 国家电网公司 | Power distribution network project construction effect Post-assessment Method based on BP neural network |
| CN110705908A (en) * | 2019-10-16 | 2020-01-17 | 海南电网有限责任公司 | County power distribution network evaluation method based on combined weighting method |
| CN112541671A (en) * | 2020-12-07 | 2021-03-23 | 国网福建省电力有限公司 | Primary and secondary fusion power distribution network construction evaluation system based on multi-stage fuzzy comprehensive evaluation |
-
2021
- 2021-04-16 CN CN202110416967.8A patent/CN113095692A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106991531A (en) * | 2017-03-30 | 2017-07-28 | 国家电网公司 | Power distribution network project construction effect Post-assessment Method based on BP neural network |
| CN110705908A (en) * | 2019-10-16 | 2020-01-17 | 海南电网有限责任公司 | County power distribution network evaluation method based on combined weighting method |
| CN112541671A (en) * | 2020-12-07 | 2021-03-23 | 国网福建省电力有限公司 | Primary and secondary fusion power distribution network construction evaluation system based on multi-stage fuzzy comprehensive evaluation |
Non-Patent Citations (2)
| Title |
|---|
| 姜洪浪: "智能电能表综合性能评级方法研究", 《电测与仪表》, 25 March 2018 (2018-03-25) * |
| 王维洲: "A+~E类区域配电网降损潜力评估模型研究", 《电力科学与技术学报》, 28 June 2018 (2018-06-28) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104951866B (en) | Line loss comprehensive management benchmarking evaluation system and method for county-level power supply enterprise | |
| CN110705873B (en) | Power distribution network running state portrait analysis method | |
| CN111582700B (en) | Method for calculating fault rate of power distribution network equipment | |
| CN107516170A (en) | A kind of difference self-healing control method based on probability of equipment failure and power networks risk | |
| CN109146124B (en) | Power distribution terminal transformation decision method based on time-varying failure rate | |
| CN114386753A (en) | Equipment risk comprehensive analysis early warning method based on main transformer load condition | |
| CN108985602B (en) | Power grid classification project input and output evaluation method and system considering risks | |
| CN104462718A (en) | Method for evaluating economic operation year range of transformer substation | |
| CN110705859A (en) | PCA-self-organizing neural network-based method for evaluating running state of medium and low voltage distribution network | |
| CN114118588A (en) | Peak-facing summer power failure prediction method based on game feature extraction under clustering undersampling | |
| CN111881961A (en) | Power distribution network fault risk grade prediction method based on data mining | |
| Dong | Combining unsupervised and supervised learning for asset class failure prediction in power systems | |
| CN112508254B (en) | Method for determining investment prediction data of transformer substation engineering project | |
| CN113327047B (en) | Power marketing service channel decision method and system based on fuzzy comprehensive model | |
| CN111178690A (en) | Electricity stealing risk assessment method for electricity consumers based on wind control scoring card model | |
| CN111091223A (en) | Distribution transformer short-term load prediction method based on Internet of things intelligent sensing technology | |
| CN113095692A (en) | Comprehensive evaluation method and system for operation quality of power distribution network circuit considering user requirements | |
| Dehghanian et al. | Optimal RTU placement in power distribution systems using a novel method based on analytical hierarchical process (AHP) | |
| CN114399150A (en) | Distribution network equipment state evaluation method based on AHP-fuzzy comprehensive evaluation method | |
| CN114692937A (en) | Main transformer load prediction method based on main transformer load influence factor prediction | |
| CN113191720A (en) | 10kV exploitable project investment decision method based on star classification and quality and efficiency evaluation | |
| CN108717597B (en) | Grid engineering operation benefit evaluation method and system for optimizing grid structure | |
| CN108492056B (en) | Method and system for evaluating operation benefits of power grid project of power supply of electric railway | |
| CN109474002B (en) | Wind power plant reliability comprehensive evaluation method considering electrical main wiring topology | |
| Falcão | Genetic algorithms applications in electrical distribution systems |
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 |