CN117977585B - Power quality assessment method for power distribution network - Google Patents
Power quality assessment method for power distribution network Download PDFInfo
- Publication number
- CN117977585B CN117977585B CN202410390360.0A CN202410390360A CN117977585B CN 117977585 B CN117977585 B CN 117977585B CN 202410390360 A CN202410390360 A CN 202410390360A CN 117977585 B CN117977585 B CN 117977585B
- Authority
- CN
- China
- Prior art keywords
- quality index
- power quality
- electric energy
- power
- index
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
- G06F18/15—Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2431—Multiple classes
-
- 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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for AC mains or AC distribution networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/80—Management or planning
- Y02P90/82—Energy audits or management systems therefor
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- General Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Primary Health Care (AREA)
- Marketing (AREA)
- Human Resources & Organizations (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention relates to the technical field of distribution networks, in particular to a power quality assessment method of a distribution network, which comprises the steps of collecting index data of the distribution network in a plurality of time periods, calculating each power quality index of each time period, obtaining a first weight value of each power quality index by using an entropy weight method, adaptively dividing all the time periods into a plurality of clustering domains by using an iterative self-organizing clustering algorithm, obtaining the superposition degree of each power quality index according to a clustering result, screening out a clustering domain to be regulated according to the superposition degree, calculating the adjustment degree of each power quality index according to the clustering domain to be regulated, regulating the first weight value of each power quality index into a second weight value by using the adjustment degree, improving the accuracy of each power quality index weight value, assessing the power quality of each time period based on the second weight value, and improving the accuracy of power quality assessment.
Description
Technical Field
The invention relates to the technical field of power distribution networks in general, and more particularly relates to a power quality evaluation method of a power distribution network.
Background
Along with the rapid development of industrial automation and information technology, the requirements of electric power users on the electric energy quality are higher and higher, the electric energy quality problem not only affects the normal operation of user equipment, but also can cause energy waste and instability of an electric power system, and in view of the fact that the electric energy quality of a power distribution network is affected by multiple complex factors, a set of scientific and strict electric energy quality assessment method with strong adaptability is needed.
The current technical means relies on collecting historical data of power indexes of a power distribution network, determining weights of all indexes by using an entropy weight method, generating a power quality index by weighting calculation, and carrying out power quality assessment based on the power quality index.
However, the entropy weight method has a certain limitation in practice, the sensitivity of the entropy weight method to data distribution is easy to cause the weight calculation result to be influenced by abnormal data or deviation, and the determined weight value cannot accurately represent the power quality index obtained by index data, so that the evaluation of the power quality of the power distribution network is not accurate enough.
Disclosure of Invention
In order to solve one or more of the above technical problems, the invention provides a power quality evaluation method for a power distribution network, which improves the accuracy of power quality evaluation by adjusting a weight determined by an entropy weight method to be a more accurate weight, and specifically adopts the following technical scheme: a power quality assessment method for a power distribution network, comprising:
Collecting various index data of the power distribution network in a plurality of time periods;
Calculating each electric energy quality index of each time period based on the index data, and determining a first weight of each electric energy quality index by utilizing an entropy weight method;
weighting each electric energy quality index according to a first weight, and clustering all time periods based on each weighted electric energy quality index to obtain a plurality of clustering domains;
obtaining the coincidence degree of each electric energy quality index in each cluster domain according to the frequency of occurrence of each electric energy quality index in each cluster domain in the range of the electric energy quality index in each cluster domain;
comparing the coincidence degree with a preset threshold value, and screening out a cluster domain to be adjusted;
adjusting the first weight of each electric energy quality index to a second weight by utilizing the superposition degree of each electric energy quality index in the cluster domain to be adjusted;
And evaluating the power quality of each time period through the second weight value of each power quality index.
Further, the adjusting the first weight of each power quality indicator to the second weight specifically includes:
Calculating the adjustment degree of each electric energy quality index in the cluster domain to be adjusted based on the superposition degree of each electric energy quality index in the cluster domain to be adjusted and the standard deviation of the electric energy quality index;
Multiplying the adjustment degree of each power quality index with the first weight of the power quality index, and taking the obtained value as the second weight of the power quality index.
Further, the adjustment degree of each electric energy quality index is calculated by the following steps:
In the method, in the process of the invention, For/>, in the cluster domain to be adjustedDegree of adjustment of item Power quality index,/>For the number of all cluster domains,/>For/>The/>, in the individual cluster domainStandard deviation of power quality index,/>For/>The/>, in the individual cluster domainThe coincidence degree of the power quality indexes.
Further, the coincidence degree of each electric energy quality index in each cluster domain specifically satisfies the following relation:
In the method, in the process of the invention, For/>Cluster domain number/>The coincidence degree of the electric energy indexes,/>For/>Number of time periods contained in each cluster domain,/>For/>Index of the number of time periods contained in the individual cluster domains,/>For/>The/>, in the individual cluster domainFirst/>, of the individual time periodsA power quality index, the/>, in each cluster domainThe number of occurrences within the range of the power quality indicator.
Further, comparing the coincidence degree with a preset threshold value, and screening out a cluster domain to be adjusted, including:
setting a preset overlap ratio threshold;
if the coincidence degree of a certain electric energy quality index in the cluster domain is larger than the coincidence degree threshold, the cluster domain is the cluster domain to be adjusted, otherwise, the cluster domain is not used as the cluster domain to be adjusted.
Further, the method includes the steps of collecting various index data of the power distribution network in a plurality of time periods, including current, voltage waveforms, frequency deviation, harmonic content and power factors.
Further, calculating each power quality indicator for each time period based on the index data comprises:
Obtaining root mean square errors of current waveforms and standard current waveforms in each time period, and taking the value normalized by the root mean square errors as a first power quality index;
Obtaining root mean square errors of the voltage waveform and the standard voltage waveform in each time period, and taking the value normalized by the root mean square errors as a second power quality index;
And respectively normalizing the frequency deviation, the harmonic content and the power factor, respectively subtracting the normalized values of the frequency deviation and the harmonic content from 1 to obtain a third power quality index and a fourth power quality index, and taking the normalized values of the power factor as a fifth power quality index.
Further, evaluating the power quality of each time period by the second weights of the power quality indicators comprises:
and carrying out weighted summation on each electric energy quality index of each time period and the second weight value of each electric energy quality index, and taking the obtained data value as the electric energy quality score of the time period.
Further, the clustering adopts an iterative self-organizing clustering algorithm.
The invention has the following effects:
According to the invention, the comprehensiveness of power quality assessment is ensured by collecting the data of each power quality index of the power distribution network in a plurality of time periods, the first weight of each index is determined by calculating each power quality index of each time period and utilizing an entropy weight method, each power quality index is weighted according to the first weight, cluster analysis is carried out to form a plurality of cluster domains, the power quality problems of different types are favorably identified and distinguished, the overlapping degree of each index is calculated by analyzing the occurrence times of each power quality index in each cluster domain, the cluster domain to be regulated, which may have deviation in weight, is identified, the first weight is adjusted to be a second weight according to the overlapping degree of each power quality index in the cluster domain to be regulated, the accuracy and the adaptability of weight distribution are enhanced, thereby the accuracy of power quality assessment is improved, the influence of the power quality of each time period and the power quality indexes of different power quality can be more accurately reflected by using the second weight after the optimization, the quality of the power quality of different time periods and the power quality problems can be favorably found and guaranteed, and the stability and the reliability of the power grid can be solved.
Drawings
The above, as well as additional purposes, features, and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description when read in conjunction with the accompanying drawings. In the drawings, embodiments of the invention are illustrated by way of example and not by way of limitation, and like reference numerals refer to similar or corresponding parts and in which:
fig. 1 is a flow chart of the method of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.
Referring to fig. 1, a power quality evaluation method for a power distribution network includes steps S1 to S7, specifically as follows:
s1: and collecting various index data of the power distribution network in a plurality of time periods.
The time of the time periods is the same, specifically, each time period is an acquisition period, the acquisition period can be set by itself, in this embodiment, the preset acquisition period is 10 minutes, that is, each item of index data in ten minutes is acquired each time, multiple times of acquisition is performed, and a special electric energy quality monitoring instrument can be used during acquisition.
Each item of index data to be collected in the power distribution network comprises data of current waveforms, voltage waveforms, frequency deviation, harmonic content and power factors. In the index data, the larger the difference between the current waveform and the standard waveform in the power distribution network is, the worse the electric energy quality is reflected, the frequency deviation is the difference between the actual value and the standard value of the system frequency of the power system under the normal operation condition, the larger the difference is, the worse the electric energy quality is, the harmonic wave can not only influence the power supply quality of the power network, and cause electric energy waste, but also heat and loss of equipment are increased, the service life is shortened, even faults or burnout occur, and loss is caused, so the worse the electric energy quality is the more the harmonic wave is, the power factor is the ratio of the active power to the apparent power in the alternating current circuit, the highest circuit efficiency is represented when the value is 1, and the power distribution network quality is evaluated by collecting the data.
S2: and calculating various power quality indexes of each time period based on the index data, and determining a first weight of the power quality indexes by using an entropy weight method.
For each index data, the value ranges of the different index data are different, the influence on the power quality is also different, five power quality indexes capable of reflecting the power quality are calculated according to the influence relation of the collected index data on the power quality, the indexes are all normalized values, the orders of magnitude of all the index data are the same as much as possible, the subsequent calculation of weights for the data of the same order of magnitude by an entropy weight method is facilitated, and therefore each power quality index smaller than 1 is obtained.
Wherein calculating each power quality index of each time period based on each index data of each time period comprises:
(1) Obtaining root mean square errors of current waveforms and standard current waveforms in each time period, taking the normalized value of the root mean square errors as a first power quality index, obtaining root mean square errors of voltage waveforms and standard voltage waveforms in each time period, and taking the normalized value of the root mean square errors as a second power quality index;
By calculating the root mean square error of the current and voltage waveforms and the standard waveform and normalizing the root mean square error, the power quality indexes of different time periods are ensured to be compared on the same order of magnitude, the influence caused by the difference of the dimension and the numerical range of the original data is avoided, and each index has comparability.
(2) And respectively normalizing the frequency deviation, the harmonic content and the power factor, respectively subtracting the normalized values of the frequency deviation and the harmonic content from 1 to obtain a third power quality index and a fourth power quality index, and taking the normalized values of the power factor as a fifth power quality index.
Because the frequency deviation, the harmonic content and the power factor may be data of different dimensions and numerical ranges originally, the data are converted to the same scale (between 0 and 1) through normalization processing, so that different indexes can be directly compared and weighted, convenience is provided for subsequent analysis and calculation, and for the two indexes of the frequency deviation and the harmonic content, generally, smaller deviation and lower harmonic content represent better power quality, therefore, by subtracting the normalized numerical value from 1, the smaller index value can be ensured, the better power quality is ensured, and conversely, the worse power quality is indicated, and the negative influence of the two indexes on the power quality can be reflected more intuitively by the design.
By the processing mode, the indexes (such as power factors, larger and better) of positive influence or the indexes (such as frequency deviation and harmonic content, smaller and better) of negative influence can be unified within a range of 0 to 1, and consistent basic data can be provided for determining the weight of each electric energy quality index by using an entropy weight method or other methods.
In summary, the above operations aim to convert complex power quality indicators into standardized and comparable forms, so as to more accurately and equitably evaluate and analyze the power quality of the power distribution network, and provide scientific basis for subsequent optimization decisions.
Wherein, the weight of each electric energy index data is obtained as a first weight according to an entropy weight method, and the first weight is obtainedThe first weight of item data is noted as/>The power quality index for each time period may then be calculated based on the first weights, including: and weighting and summing each electric energy quality index of each time period and the corresponding first weight value, and taking the obtained value as the electric energy quality index of the time period.
It should be noted that, the power quality index of all the time periods in each cluster domain is obtained in this way, the power quality index can comprehensively reflect the power quality of the time period, and the higher the power quality index is, the better the power quality of the time period is, but the lower the accuracy of the weight determined by the entropy weight method is, the evaluation of the power quality is affected, so the subsequent analysis is performed to determine the accurate weight.
S3: and weighting each electric energy quality index according to a first weight, and clustering all time periods based on each weighted electric energy quality index to obtain a plurality of clustering domains.
After each electric energy quality index of each time period is weighted with the first weight thereof, each electric energy quality index of all time periods weighted according to each electric energy quality index is adaptively classified into a plurality of clustering domains by an iterative self-organizing clustering algorithm:
Since the entropy weight method considers the relative importance of each power quality index in the overall influence when calculating the first weight, the entropy weight method only considers the relevance and the joint effect among different indexes from the single index point of view. In view of the fact that the importance of each item of electric energy quality index data to electric energy quality assessment is different, an iterative self-organizing clustering algorithm is adopted to carry out self-adaptive classification on each item of electric energy quality index after weighing in all time periods, and a data set is divided into a plurality of clustering domains.
Specifically, first weighting a first weight corresponding to each power quality index of each time period, namely multiplying each power quality index of each time period by the first weight corresponding to the index to obtain each power quality index weighted by all time periods, then comprehensively analyzing all time periods by using an iterative self-organizing clustering algorithm, and automatically classifying the time periods into different clustering domains according to the similarity of the weighted index values, wherein the aim of the link is to feed back according to actual classification results, provide basis for further adjusting and optimizing the first weights of each power quality index, and ensure the accuracy and the comprehensiveness of power quality assessment.
The iterative self-organizing clustering algorithm is an improvement on the K-means algorithm, can automatically generate a plurality of high-quality classifications without manually setting the number of class clusters to be generated, generates a plurality of clustering domains, each clustering domain comprises a plurality of time periods, each time period corresponds to five power quality indexes, and the acquired number of the clustering domains is recorded asWill/>The number of time periods contained in the individual cluster domains is noted/>。
S4: and obtaining the coincidence degree of each electric energy quality index in each cluster domain according to the occurrence frequency of each electric energy quality index in each cluster domain in the range of the electric energy quality index in each cluster domain.
The coincidence degree of each electric energy quality index in each cluster domain meets the following relation:
In the method, in the process of the invention, For/>Cluster domain number/>The coincidence degree of the electric energy indexes,/>For/>Number of time periods contained in each cluster domain,/>For/>Index of the number of time periods contained in the individual cluster domains,/>For/>The/>, in the individual cluster domainFirst/>, of the individual time periodsA power quality index, the/>, in all cluster domainsThe number of occurrences within the range of the power quality indicator.
Reflecting the/>, in the cluster domainCluster domain number/>The overlapping degree of the item electric energy index and the index range in other cluster domains,/>For measuring/>First/>, of the individual time periodsHow much the item electric energy index overlaps with the same item index range in other cluster domains, and the whole calculated average value is used for measuring the coincidence degree of the item index in the cluster domain compared with other cluster domains, and the high coincidence degree indicates the/>First/>, of the clustering domainsThe degree of distinction between the term index and other clustering domains is not high, the first weight value may need to be adjusted to improve the clustering effect and accuracy, and through the formula, the similarity of the power quality index inside each clustering domain and other clustering domains can be quantitatively analyzed, so that the clustering domain which may need to be adjusted in weight value is identified, and the power quality assessment and classification result is further optimized.
According to the formula, the accuracy and rationality of a clustering result are evaluated by comparing the index distribution condition of the inner indexes of the clustering domains with the index distribution condition of the whole range of all the clustering domains, in an ideal clustering result, data in the same clustering domain should be highly similar, and obvious differences exist among different clustering domains, however, as the interaction and influence among various power quality indexes can not be fully reflected by a first weight, the power quality indexes among different clustering domains can be excessively close, namely, the data overlap ratio is higher, and for any one clustering domain, the higher the overlap ratio is, the similarity of the power quality index of the clustering domain and other clustering domains is larger, which means that the original weight setting is possibly insufficient to effectively distinguish different power quality grades, so that the clustering domains to be adjusted, namely, the clustering domains with the higher overlap ratio, can be screened out by calculating the overlap ratio, so that the clustering result is optimized, the similarity of the data in the classes is ensured, the data difference between the classes is obvious, and the accuracy and the effectiveness of the power quality evaluation are improved.
In summary, the formula aims to evaluate and optimize the weight distribution of the power quality index according to the clustering result so as to realize more accurate and effective power quality clustering analysis.
Wherein, each power quality index of each time period, the frequency of occurrence in the range of the power quality index in each cluster domain is obtained by the following steps:
Obtaining the maximum and minimum values of each power quality index in all time periods of each cluster domain, and obtaining the range of each power quality index in each cluster domain;
counting the number of occurrences of the power quality indicator in the range of the power quality indicator in each cluster domain, a specific example being:
Such as the total cluster domain comprises a cluster domain A and a cluster domain B;
the clustering domain A comprises a time period 1 and a time period 2, wherein the first power quality index of the time period 1 is 0.3, and the first power quality index of the time period 2 is 0.4, and the range of the first power quality index in the clustering domain A is 0.3 to 0.4;
The clustering domain B comprises a time period 3 and a time period 4, wherein the first power quality index of the time period 3 is 0.3, the first power quality index of the time period 4 is 0.5, and the range of the first power quality index in the clustering domain B is 0.3 to 0.5;
the first power quality index of the period 1 is 0.3 within the range of 0.3 to 0.4 of the first power quality index in the cluster domain a and also within the range of 0.3 to 0.5 of the first power quality index in the cluster domain B, and thus the number of occurrences of the first power quality index of the period 1 in each cluster domain is 2.
S5: and comparing the coincidence degree with a preset threshold value, and screening out a cluster domain to be adjusted.
The coincidence degree of each power quality index in each cluster domain and the power quality index in each time period are simultaneously compared with a preset coincidence degree threshold value so as to carry out screening.
The screening of the cluster domain to be adjusted comprises the following steps:
The preset overlap ratio threshold is set, the experience value is set to be 1.2 in the embodiment, and the embodiment can be specifically set by oneself;
If the electric energy quality index with the overlapping degree larger than the preset overlapping degree threshold value exists in the clustering domain or the time period with the electric energy quality index larger than the preset overlapping degree threshold value exists in the clustering domain, the clustering domain is the clustering domain to be adjusted, otherwise, the clustering domain is not the clustering domain to be adjusted.
S6: and adjusting the first weight of each power quality index to a second weight by utilizing the coincidence degree of each power quality index in the cluster domain to be adjusted.
The method comprises the following steps of:
S61: calculating the adjustment degree of each electric energy quality index in the cluster to be adjusted based on the superposition degree of each electric energy quality index in the cluster to be adjusted and the standard deviation of the electric energy quality index, wherein the adjustment degree specifically meets the following relation:
In the method, in the process of the invention, For/>, in the cluster domain to be adjustedDegree of adjustment of item Power quality index,/>For the number of all cluster domains,/>For/>The/>, in the individual cluster domainStandard deviation of power quality index,/>For/>Cluster domain number/>The coincidence degree of the power quality indexes.
Further, the degree of coincidence in the formula reflects the distribution similarity of a certain electric energy quality index among different clustering domains, whether the electric energy quality index among different clustering domains is difficult to distinguish due to weight setting can be reflected, the standard deviation measures the data discrete degree of the certain electric energy quality index in a single clustering domain, the degree of coincidence and the standard deviation indirectly reflect the sensitivity of the index to electric energy quality change, the degree of coincidence is complementary, the degree of coincidence focuses on the similarity problem of cross domains, the latter focuses on the fluctuation of data in the domains, and the weight of the electric energy quality index can be more comprehensively estimated and adjusted by combining the degree of coincidence and the standard deviation.
When the overlap ratio is higher, the setting of the weight value may not effectively distinguish different electric energy quality levels, and at the moment, if the standard deviation is also larger, the index has large fluctuation range in the clustering domain, and weight value adjustment is needed to be carried out on the index, otherwise, if the overlap ratio is low and the standard deviation is also small, the index has better distinction between different clustering domains, and the data in the clustering domain is more concentrated, and the weight value may not need to be adjusted greatly.
The standard deviation is used as a statistic for describing the degree of dispersion of data, can be used for reflecting the variability and uncertainty of the data, is combined with the concept of the degree of coincidence in cluster analysis, can provide a weight adjustment strategy which is more in line with reality in the power quality evaluation, and the formula skillfully balances the influence of the degree of dispersion and the degree of coincidence by combining the standard deviation with an exponential form of the degree of coincidence subtracted by 1, reduces the influence of the standard deviation on the degree of coincidence when the degree of coincidence is higher, amplifies the influence of the standard deviation when the degree of coincidence is lower, and promotes the weight adjustment of the indexes to be larger in amplitude.
According to the adjustment degree, the initial weight is appropriately increased or decreased, and for the indexes with high overlap ratio and large standard deviation, the weight is reduced to reduce the data range, so that confusion of the power quality indexes among different clustering domains is reduced, the clustering distinguishing capability is improved, and the accuracy of the evaluation result is ensured.
In summary, the combination of the overlap ratio and the standard deviation has the advantages that the overall view angle and the local view angle can be considered, the similarity of the power quality indexes among different clustering domains is considered, and the fluctuation condition of the data in a single clustering domain is considered, so that the optimization adjustment of the power quality index weight is guided more accurately, and the evaluation precision and effectiveness are improved.
S62: the value obtained by multiplying the adjustment degree of each electric energy quality index and the first weight of the electric energy quality index is used as the second weight of the electric energy quality index, and the value is expressed as follows by a formula:
In the method, in the process of the invention, Represents the/>A second weight of the power quality index,/>Represents the/>Degree of adjustment of item Power quality index,/>Represents the/>A first weight of the power quality indicator.
In order to improve the possible defects of the first weight obtained based on the entropy weight method, namely the equality of the power quality indexes of all time periods in the same clustering domain and the obvious distinction of the power quality indexes among different clustering domains cannot be ensured, the second weight of the power quality indexes is introduced, the first weight is adjusted to be the second weight, the aim is to enable the power quality index levels calculated by weighting according to the second weight of the time periods belonging to the same clustering domain to be closer, and for the time periods of different clustering domains, the power quality indexes have larger differences, so that the clustering result is more consistent with the actual power quality condition, the adjustment process aims to strengthen the consistency of the data in the clustering domain and the difference of the data among the clustering domains, so as to realize accurate clustering effect, ensure the accuracy of power quality evaluation, and in a word, the actual influence degree of the power quality indexes among different time periods and the clustering domains can be more accurately reflected by adjusting the first weight to be the second weight, and the reliability and the practicability of the power quality evaluation result are further improved.
S7: and evaluating the power quality of each time period through the second weight value of each power quality index.
The specific evaluation method comprises the following steps: and weighting and summing each power quality index of each time period according to the corresponding second weight value of each time period, taking the obtained data value as the power quality score of the time period, wherein the higher the score is, the better the power quality of the power distribution network is, and the power quality evaluation of the power distribution network can be operated according to the steps of the method of S1-S7, so that the accurate evaluation of the power quality of the power distribution network is realized.
In the description of the present specification, the meaning of "a plurality", "a number" or "a plurality" is at least two, for example, two, three or more, etc., unless explicitly defined otherwise.
While various embodiments of the present invention have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. Many modifications, changes, and substitutions will now occur to those skilled in the art without departing from the spirit and scope of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in practicing the invention.
Claims (6)
1. A power quality assessment method for a power distribution network, comprising:
Collecting various index data of the power distribution network in a plurality of time periods;
Calculating each electric energy quality index of each time period based on the index data, and determining a first weight of each electric energy quality index by utilizing an entropy weight method;
weighting each electric energy quality index according to a first weight, and clustering all time periods based on each weighted electric energy quality index to obtain a plurality of clustering domains;
obtaining the coincidence degree of each electric energy quality index in each cluster domain according to the frequency of occurrence of each electric energy quality index in each cluster domain in the range of the electric energy quality index in each cluster domain;
comparing the coincidence degree with a preset threshold value, and screening out a cluster domain to be adjusted;
adjusting the first weight of each electric energy quality index to a second weight by utilizing the superposition degree of each electric energy quality index in the cluster domain to be adjusted;
evaluating the power quality of each time period through the second weight of each power quality index;
The adjusting the first weight of each power quality index to the second weight specifically includes:
Calculating the adjustment degree of each electric energy quality index in the cluster domain to be adjusted based on the superposition degree of each electric energy quality index in the cluster domain to be adjusted and the standard deviation of the electric energy quality index;
Multiplying the adjustment degree of each electric energy quality index with the first weight of the electric energy quality index, and taking the obtained value as the second weight of the electric energy quality index;
The adjustment degree of each electric energy quality index is calculated by the following steps:
In the method, in the process of the invention, For/>, in the cluster domain to be adjustedDegree of adjustment of item Power quality index,/>For the number of all cluster domains,For/>The/>, in the individual cluster domainStandard deviation of power quality index,/>For/>The/>, in the individual cluster domainThe degree of coincidence of the power quality indexes;
the coincidence degree of each electric energy quality index in each cluster domain specifically meets the following relation:
In the method, in the process of the invention, For/>Cluster domain number/>The coincidence degree of the electric energy indexes,/>For/>Number of time periods contained in each cluster domain,/>For/>Index of the number of time periods contained in the individual cluster domains,/>For/>The/>, in the individual cluster domainFirst/>, of the individual time periodsA power quality indicator, a number of occurrences of the power quality indicator in each cluster domain.
2. The method for evaluating the power quality of a power distribution network according to claim 1, wherein comparing the overlapping degree with a preset threshold value, screening out a cluster domain to be adjusted, comprises:
setting a preset overlap ratio threshold;
if the coincidence degree of a certain electric energy quality index in the cluster domain is larger than the coincidence degree threshold, the cluster domain is the cluster domain to be adjusted, otherwise, the cluster domain is not used as the cluster domain to be adjusted.
3. The method for evaluating the power quality of a power distribution network according to claim 1, wherein the collecting of each index data of the power distribution network in a plurality of time periods includes current, voltage waveform, frequency deviation, harmonic content and power factor.
4. A power distribution network power quality assessment method according to claim 3, wherein calculating each power quality index for each time period based on said index data comprises:
Obtaining root mean square errors of current waveforms and standard current waveforms in each time period, and taking the value normalized by the root mean square errors as a first power quality index;
Obtaining root mean square errors of the voltage waveform and the standard voltage waveform in each time period, and taking the value normalized by the root mean square errors as a second power quality index;
And respectively normalizing the frequency deviation, the harmonic content and the power factor, respectively subtracting the normalized values of the frequency deviation and the harmonic content from 1 to obtain a third power quality index and a fourth power quality index, and taking the normalized values of the power factor as a fifth power quality index.
5. A power distribution network power quality assessment method according to claim 1, wherein assessing the power quality of each time period by the second weights of the power quality indicators comprises:
and carrying out weighted summation on each electric energy quality index of each time period and the second weight value of each electric energy quality index, and taking the obtained data value as the electric energy quality score of the time period.
6. The method for evaluating the power quality of a power distribution network according to claim 1, wherein the clustering adopts an iterative self-organizing clustering algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410390360.0A CN117977585B (en) | 2024-04-02 | 2024-04-02 | Power quality assessment method for power distribution network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410390360.0A CN117977585B (en) | 2024-04-02 | 2024-04-02 | Power quality assessment method for power distribution network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117977585A CN117977585A (en) | 2024-05-03 |
| CN117977585B true CN117977585B (en) | 2024-06-07 |
Family
ID=90866129
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410390360.0A Active CN117977585B (en) | 2024-04-02 | 2024-04-02 | Power quality assessment method for power distribution network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117977585B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119359166B (en) * | 2024-12-24 | 2025-03-07 | 陕西佰特睿驰新能源科技有限公司 | Lithium carbonate production quality data analysis method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105894034A (en) * | 2016-04-01 | 2016-08-24 | 中国电力科学研究院 | Commercial load characteristic clustering analysis method based on quadratic clustering |
| CN109389282A (en) * | 2018-08-17 | 2019-02-26 | 浙江华云信息科技有限公司 | A kind of electric energy meter production firm evaluation method based on gauss hybrid models |
| CN112994015A (en) * | 2021-02-26 | 2021-06-18 | 国网江苏省电力有限公司海安市供电分公司 | Voltage sag comprehensive treatment method based on sensitive load clustering |
| CN114895137A (en) * | 2022-06-29 | 2022-08-12 | 广东电网有限责任公司佛山供电局 | Power supply quality testing method for power supply equipment |
| CN114926085A (en) * | 2022-06-13 | 2022-08-19 | 国网湖南省电力有限公司 | A method and system for evaluating the adjustment capability of thermal power units based on data-driven |
| CN115936482A (en) * | 2022-11-02 | 2023-04-07 | 国网浙江省电力有限公司桐乡市供电公司 | Low-voltage distribution network power quality analysis method based on power consumption data |
| CN117220281A (en) * | 2023-09-15 | 2023-12-12 | 广东电网有限责任公司 | Electric automobile access power grid adjustment capability quantitative evaluation method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11221353B2 (en) * | 2018-07-06 | 2022-01-11 | Schneider Electric USA, Inc. | Systems and methods for analyzing power quality events in an electrical system |
-
2024
- 2024-04-02 CN CN202410390360.0A patent/CN117977585B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105894034A (en) * | 2016-04-01 | 2016-08-24 | 中国电力科学研究院 | Commercial load characteristic clustering analysis method based on quadratic clustering |
| CN109389282A (en) * | 2018-08-17 | 2019-02-26 | 浙江华云信息科技有限公司 | A kind of electric energy meter production firm evaluation method based on gauss hybrid models |
| CN112994015A (en) * | 2021-02-26 | 2021-06-18 | 国网江苏省电力有限公司海安市供电分公司 | Voltage sag comprehensive treatment method based on sensitive load clustering |
| CN114926085A (en) * | 2022-06-13 | 2022-08-19 | 国网湖南省电力有限公司 | A method and system for evaluating the adjustment capability of thermal power units based on data-driven |
| CN114895137A (en) * | 2022-06-29 | 2022-08-12 | 广东电网有限责任公司佛山供电局 | Power supply quality testing method for power supply equipment |
| CN115936482A (en) * | 2022-11-02 | 2023-04-07 | 国网浙江省电力有限公司桐乡市供电公司 | Low-voltage distribution network power quality analysis method based on power consumption data |
| CN117220281A (en) * | 2023-09-15 | 2023-12-12 | 广东电网有限责任公司 | Electric automobile access power grid adjustment capability quantitative evaluation method and system |
Non-Patent Citations (1)
| Title |
|---|
| 计及模糊聚类的配电网可靠性多指标定价方法;陈碧云;郑瑜;;电网与清洁能源;20181125(第11期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117977585A (en) | 2024-05-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107609783B (en) | Method and system for evaluating comprehensive performance of intelligent electric energy meter based on data mining | |
| TWI440862B (en) | Electrical detection method and system based on user feedback information | |
| CN112381476B (en) | Method and apparatus for determining abnormal state of electric energy meter | |
| CN110221976B (en) | Quantitative evaluation method for quality of metering terminal software based on measurement technology | |
| CN111949939B (en) | Evaluation method of smart meter operating state based on improved TOPSIS and cluster analysis | |
| CN117977585B (en) | Power quality assessment method for power distribution network | |
| CN112001644B (en) | Method, device, terminal and storage medium for detecting operation reliability of power distribution network | |
| CN113125903A (en) | Line loss anomaly detection method, device, equipment and computer-readable storage medium | |
| CN117992776B (en) | Real-time prediction method for health state of power grid equipment based on artificial intelligence | |
| CN111242430A (en) | Power equipment supplier evaluation method and device | |
| CN117435937A (en) | Smart electric meter abnormal data identification method, device, equipment and storage medium | |
| CN119167218A (en) | A risk identification method based on decision tree algorithm | |
| CN118585944B (en) | A method for determining abnormalities in hazardous waste management data | |
| CN113902565A (en) | Risk assessment method, device, equipment and storage medium for financial products | |
| CN118735703A (en) | Enterprise financial data deviation prediction method and system | |
| CN118194202A (en) | Electricity theft identification algorithm based on horizontal federation and its prototype system | |
| CN115278706B (en) | A network structure evaluation method, device, equipment and computer storage medium | |
| CN114139408A (en) | Power transformer health state assessment method | |
| CN119247243B (en) | Method and device for enhancing precision of electricity acquisition data | |
| CN118917390B (en) | Service knowledge base management system and method based on knowledge big model | |
| CN119357874B (en) | An online calibration control method for a smart meter communication module | |
| CN119471544A (en) | An improved AdaBoost algorithm for fault analysis of electric energy meter measurement data | |
| CN118890568A (en) | A data correction method and system based on centralized meter reading | |
| CN119621518A (en) | CPU load stability evaluation method for multiple time scales in test field | |
| CN118966813A (en) | A method, device, electronic equipment and storage medium for automatically locating high-loss transmission area |
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 | ||
| GR01 | Patent grant |