CN115481786A - Power distribution network user load characteristic classification method based on data acquisition - Google Patents
Power distribution network user load characteristic classification method based on data acquisition Download PDFInfo
- Publication number
- CN115481786A CN115481786A CN202211045414.7A CN202211045414A CN115481786A CN 115481786 A CN115481786 A CN 115481786A CN 202211045414 A CN202211045414 A CN 202211045414A CN 115481786 A CN115481786 A CN 115481786A
- Authority
- CN
- China
- Prior art keywords
- user load
- data
- load
- user
- load characteristic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 26
- 238000012935 Averaging Methods 0.000 claims abstract description 6
- 238000000605 extraction Methods 0.000 claims abstract description 6
- 238000007621 cluster analysis Methods 0.000 claims abstract description 4
- 230000009467 reduction Effects 0.000 claims abstract description 4
- 238000012216 screening Methods 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 22
- 238000012545 processing Methods 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000005611 electricity Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 230000007547 defect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012706 support-vector machine 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- 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
- H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Power Engineering (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a power distribution network user load characteristic classification method based on data acquisition, which comprises the following steps: 1. constructing an original user load characteristic database by using user data acquired by an acquisition system, and normalizing the data to be unified into a rough numerical value interval; 2. carrying out feature extraction on an original user load feature database by using a US-ELM algorithm, screening out load features which have the most representative line on user load characteristics, and constructing a user load feature set; 3. introducing an improved FCM algorithm to perform cluster analysis on the user load feature set, taking the low-dimensional data subjected to dimension reduction in the step 2 as the input of the improved FCM algorithm, and clustering the user load feature set into a C class; 4. and classifying the original user load data by using the result obtained by clustering, and averaging the user load data of the same class to obtain a typical load curve representing different user load characteristics. The invention is beneficial to the whole load structure of the region under jurisdiction of the power supply part.
Description
Technical Field
The invention belongs to the technical field of power grid regulation and control operation, and particularly relates to a power distribution network user load characteristic classification method based on data acquisition.
Background
With continuous deepening of market reformation of the power company and continuous improvement of service awareness of users, the power company scientifically classifies the load characteristics of the power distribution network users and summarizes corresponding typical load curves, is beneficial to the power company to integrally master the load operation condition of the power distribution network and carry out work for specific users, can further carry out work such as power grid scheduling, load prediction, ordered power utilization, bearing capacity evaluation and the like according to the power utilization type and topological relation of the load, and provides a precondition for safe and stable operation analysis of the power distribution network.
The power distribution network is used as an important component of a power grid, is directly oriented to power consumers, is closely related to production life, is an important infrastructure for guaranteeing and improving the livelihood, and is also the most intuitive object for the users to experience and experience the power grid service, the scale of the power distribution network in the modern city is continuously increased, and various users are continuously accessed, so that the load characteristics of the power distribution network are more and more diversified and differentiated. Different types of user load characteristics are affected by a variety of factors, such as season, region, industry characteristics, holidays, electrical equipment, and the like. Currently, most of the mastery of the categories of the power distribution network users is based on the power utilization types provided by the users to the power company during installation, and the actual power utilization load characteristics have certain difference in time sequence change from the power utilization types reported during installation. With the continuous development of intelligent terminal technology, the power load condition of a user can be monitored, and meanwhile, the user load data collected by the power distribution network also grows exponentially, wherein complicated load information such as information loss, much noise and the like is mixed. In the face of the current situation, it is very difficult to select user categories one by using an artificial statistical method, and a more efficient and accurate method is needed to classify the load characteristics of the power distribution network users. The big data technology is excellent in processing capacity for massive data and complex data, and a new solution is provided for load characteristic classification work of the power distribution network.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a power distribution network user load characteristic classification method based on data acquisition.
The above purpose of the invention is realized by the following technical scheme:
a power distribution network user load characteristic classification method based on data acquisition is characterized by comprising the following steps: the method comprises the following steps:
step 2, carrying out feature extraction on an original user load feature database by using an US-ELM algorithm, screening out load features which have the most representative rows for user load features, constructing a user load feature set, and effectively reducing the dimensionality of the original user load feature database;
step 3, introducing an improved FCM algorithm to perform cluster analysis on the user load feature set, taking the low-dimensional data subjected to dimension reduction in the step 2 as the input of the improved FCM algorithm, and clustering the user load feature set into a C type;
and 4, carrying out classification on the original user load data by using the result obtained by clustering, and carrying out averaging processing on the user load data of the same class to obtain a typical load curve representing different user load characteristics.
Further: the step 1 comprises the following steps:
1.1 building original user load characteristic database
In a power distribution network, a data set of each user load characteristic acquired by an acquisition system is X, and X = { X =isset 1 ,x 2 ,…,x m The method comprises the steps that (1) the set of all load characteristic information is set, wherein the set comprises current, active power, reactive power, electricity consumption and the like; let D be the original user load characteristic database, which contains load characteristic sets X of n users, and D can be represented as:
1.2 normalizing the data in the original user load characteristic database
Because the original user load characteristic database contains data points of different categories and dimensions, and simultaneously, singular sample data may exist, the training time is increased, and the result that convergence cannot be achieved is caused. In order to avoid this situation and the convenience of subsequent data processing, and to accelerate the network learning speed, the data in the database needs to be normalized as shown in the following formula:
wherein x is ij Set of load characteristics X for the ith user i Data point of (1), x ij ∈D,x imax Set of load characteristics X for the ith user i The maximum value is the load characteristic set X of the ith user i The normalized original user load characteristic database may be represented as:
further: the step 2 comprises the following steps:
2.1 randomly selecting initial values of an input weight vector w and a deviation vector b by using an Analytic Hierarchy Process (AHP);
2.2 calculate the output matrix H as follows:
wherein g (-) is an activation function, and sigmod (-) function, x' ij Load characteristic data points in D'.
2.3 calculate the Laplace matrix L as follows:
L=D-W
wherein n is the number of users, and the calculation formulas of W and D are respectively:
2.4, solving the following formula to obtain an output weight value beta:
min||β|| 2 +λTr(β T H T LHβ),s.t.(Hβ) T Hβ=I
wherein λ represents a compromise parameter, tr (·) represents the trajectory of the matrix, and to avoid the degenerate solution of β, the above equation can be replaced by the generalized eigenvalue problem:
(I+λH T LH)v=γH T Hv
wherein, I is a unit matrix, v is a characteristic vector, and gamma is a characteristic value. Solving the above equation to obtain l eigenvectors, the output weight β will be composed of l generalized eigenvectors v, that is:
2.5 after obtaining the output weight beta, calculating a low-dimensional output matrix E through the following formula, namely mapping the input data to a user load characteristic set of an l-dimensional characteristic space through an US-ELM algorithm;
and further: the step 3 comprises the following steps:
step 3.1, testing different types of numbers by using DB as a judgment standard in an initial stage, determining the type number C, and setting other flexible parameters of a control algorithm, including an initial membership matrix u 0 A weighting index q;
step 3.2, calculating a clustering center a:
where p is the number of iterations, e k For a data set in E, u ik Represents the ith cluster center and the kth data e k A membership matrix between;
step 3.3, modifying the membership matrix u:
step 3.4 whenAnd stopping iteration, otherwise, executing the step 3.2, and performing the next iteration, wherein the iteration times p = p +1 until the errors of the two membership matrixes are within a specified range, and clustering the user load feature set into a C class.
Step 3.5, the membership degrees of the cluster centers of the user load feature sets in each class are ranked, and the user load feature set v with high membership degree is selected i (i =1,2,.., h) and clustering result label t i (I =1,2.,. H) are combined into training samples I for a vector machine (SVM) classifier, namely:
and 3.6, classifying the residual user load feature set with the low membership value by using the trained SVM classifier to obtain a classification label, and replacing the clustering label of the residual user load feature set with the low membership value to obtain more accurate classification of the user load feature set by the improved FCM algorithm.
Further: the averaging process in step 4 is shown in the following formula:
wherein, Y i The typical load data set after being homogenized for the ith class of users, p is the number of users contained in the ith class, and x ij Load data points for the ith user are shown.
The invention has the advantages and positive effects that:
the invention provides a power distribution network user load characteristic classification method based on collected data, which comprises the steps of firstly carrying out normalization processing on a massive original user load characteristic database, overcoming the defects of long training time and incapability of convergence, then carrying out characteristic extraction by using US-ELM (US-ELM), obtaining a user load characteristic set which is most representative of the user load characteristic, effectively compressing the dimensionality of an input set of a clustering algorithm, then clustering the user load characteristic set by using an improved Fuzzy C Mean (FCM) algorithm, dividing the user load characteristic into different categories, sorting the internal membership of the different categories according to the degree of membership, combining the user load characteristic set with high membership and the clustering label as the input of a Support Vector Machine (SVM) for training, classifying the user load characteristic set with low membership by using the trained SVM, updating the category label of the user load characteristic set with low membership by using a classification result, quickly and effectively classifying all users on the basis of the power distribution network user data, then carrying out uniform processing on the user load data of the same category to obtain a typical load characteristic set of the user load of the low membership by using the classification result, thereby being beneficial to the effective load characteristic of power distribution network power supply and carrying out the overall load scheduling and accurate power distribution network load and the like.
Drawings
Fig. 1 is a flow chart of a method for classifying load characteristics of a power distribution network based on data acquisition according to the present invention.
Detailed Description
The structure of the present invention will be further described by way of examples with reference to the accompanying drawings. It is to be understood that this embodiment is illustrative and not restrictive.
Referring to fig. 1, a method for classifying load characteristics of a power distribution network based on data acquisition includes the following steps:
1.1 building original user load characteristic database
In a power distribution network, a data set of load characteristics of each user acquired by a system is X, and X = { X = (X) } 1 ,x 2 ,…,x m The is the set of all load characteristic information, including current, active power, reactive power, electricity usage, etc. Let D be the original user load characteristic database, which contains load characteristic sets X of n users, and D can be represented as:
1.2 normalizing the data in the original user load characteristic database
Because the original user load characteristic database contains data points of different categories and dimensions, and simultaneously, singular sample data may exist, the training time is increased, and the result that convergence cannot be achieved is caused. In order to avoid this situation and the convenience of subsequent data processing, and to accelerate the network learning speed, the data in the database needs to be normalized as shown in the following formula:
wherein x is ij Set of load characteristics X for the ith user i Data point of (1), x ij ∈D,x imax Set of load characteristics X for the ith user i The maximum value is the load characteristic set X of the ith user i The normalized original user load characteristic database may be represented as:
step 2, carrying out feature extraction on an original user load feature database by using an US-ELM (unsupervised extreme learning machine) algorithm, screening out load features which most represent user load characteristics, constructing a user load feature set, effectively reducing the dimensionality of the original user load feature database, and providing a data basis for subsequent classification work; the method specifically comprises the following steps:
taking the normalized load database D' as input, mapping the high-dimensional load data to a low-dimensional space by using an US-ELM algorithm, and realizing the feature extraction of the original user load characteristic database, wherein the specific steps are as follows:
2.1 randomly selecting initial values of an input weight vector w and a deviation vector b by using an Analytic Hierarchy Process (AHP);
2.2 calculate the output matrix H as follows:
wherein g (-) is an activation function, and sigmod (-) function, x' ij Load characteristic data points in D'.
2.3 calculate the Laplace matrix L as follows:
L=D-W
wherein n is the number of users, and the calculation formulas of W and D are respectively:
2.4, solving the following formula to obtain an output weight value beta:
min||β|| 2 +λTr(β T H T LHβ),s.t.(Hβ) T Hβ=I
where λ represents a compromise parameter and Tr (-) represents the trajectory of the matrix, to avoid a degenerate solution of β, the above equation can be replaced with a generalized eigenvalue problem:
(I+λH T LH)v=γH T Hv
wherein, I is a unit matrix, v is a characteristic vector, and gamma is a characteristic value. Solving the above equation to obtain l eigenvectors, the output weight β will be composed of l generalized eigenvectors v, that is:
2.5 obtaining the output weight beta, calculating a low-dimensional output matrix E by the following formula, namely mapping the input data to the user load feature set of the l-dimensional feature space by the US-ELM algorithm.
And 3, introducing an improved fuzzy C-means (FCM) algorithm to perform cluster analysis on the user load feature set, taking the low-dimensional data subjected to the dimension reduction of the US-ELM as the input of the improved FCM algorithm, and clustering the user load feature set into a C class. The method specifically comprises the following steps:
step 3.1, testing different types of numbers by using Davies-Bouldin (DB) as a judgment standard in an initial stage, determining the type number C, and setting other flexible parameters of a control algorithm, including an initial membership matrix u 0 A weighting index q;
step 3.2, calculating clustering center a
Where p is the number of iterations, e k Is a data set in E, u ik Represents the ith cluster center and the kth data e k C is a clustering number;
step 3.3, modifying the membership matrix u:
step 3.4 whenAnd stopping iteration, otherwise, executing the step 3.2, and performing the next iteration, wherein the iteration times p = p +1 until the errors of the two membership matrixes are within a specified range, and clustering the user load feature set into a C class.
Step 3.5, the user load feature sets in each class are ranked according to the membership degree of the clustering center, and the user load feature set v with high membership degree is selected i (i =1,2,.., h) and clustering result label t i (I =1,2,.. H) are combined into training samples I for a vector machine (SVM) classifier, namely:
step 3.6, classifying the residual user load feature set with low membership value by using the trained SVM classifier to obtain a classification label, replacing the clustering label of the residual user load feature set with low membership value to obtain more accurate classification of the user load feature set by the improved FCM algorithm, and comparing the classification label with the standard FCM algorithm and the k-means algorithm, wherein the comparison result refers to the following table:
step 4, carrying out category division on the original user load data by using the result obtained by clustering, and carrying out averaging processing on the user load data of the same category, wherein the following formula is shown as follows:
wherein, Y i The typical load data set after being homogenized for the ith class of users, p is the number of users contained in the ith class, and x ij Load data points for the ith user of the ith class.
And finally, obtaining typical load curves representing different user load characteristics according to the average value of the user load data of different classes.
Although the embodiments of the present invention and the accompanying drawings are disclosed for illustrative purposes, those skilled in the art will appreciate that: various substitutions, changes and modifications are possible without departing from the spirit of the invention and the scope of the appended claims, and therefore the scope of the invention is not limited to the disclosure of the embodiments and the accompanying drawings.
Claims (5)
1. A power distribution network user load characteristic classification method based on data acquisition is characterized by comprising the following steps: the method comprises the following steps:
step 1, constructing an original user load characteristic database by using user data acquired by an acquisition system, and carrying out normalization processing on the data in the original user load characteristic database to unify the data into a rough numerical value interval;
step 2, carrying out feature extraction on an original user load feature database by using an US-ELM algorithm, screening out load features which have the most representative rows for user load features, constructing a user load feature set, and effectively reducing the dimensionality of the original user load feature database;
step 3, introducing an improved FCM algorithm to perform cluster analysis on the user load feature set, taking the low-dimensional data subjected to dimension reduction in the step 2 as the input of the improved FCM algorithm, and clustering the user load feature set into a C type;
and 4, classifying the original user load data by using the clustering result, and averaging the user load data of the same class to obtain typical load curves representing different user load characteristics.
2. The method for classifying user load characteristics of a power distribution network based on data acquisition according to claim 1, wherein the method comprises the following steps: the step 1 comprises the following steps:
1.1 building original user load characteristic database
In a power distribution network, a data set of load characteristics of each user acquired by a system is X, and X = { X = (X) } 1 ,x 2 ,…,x m The method comprises the steps that (1) the set of all load characteristic information is set, wherein the set comprises current, active power, reactive power, electricity consumption and the like; let D be the original user load characteristic database, which contains load characteristic sets X of n users, and D can be represented as:
1.2 normalizing the data in the original user load characteristic database
Because the original user load characteristic database contains data points of different categories and dimensions, and simultaneously, singular sample data may exist, the training time is increased and the result of convergence is not possible; in order to avoid this situation and the convenience of subsequent data processing, and to accelerate the network learning speed, the data in the database needs to be normalized as shown in the following formula:
wherein x is ij Set of load characteristics X for the ith user i Data point of (1), x ij ∈D,x imax Set of load characteristics X for the ith user i The maximum value of the load characteristic set X is the ith user i The normalized original user load characteristic database may be represented as:
3. the method for classifying user load characteristics of a power distribution network based on data acquisition according to claim 2, wherein the method comprises the following steps: the step 2 comprises the following steps:
2.1 randomly selecting initial values of an input weight vector w and a deviation vector b by using an Analytic Hierarchy Process (AHP);
2.2 calculate the output matrix H as follows:
wherein g (-) is an activation function, and sigmod (-) function, x' ij Load characteristic data points in D';
2.3 calculate the Laplace matrix L as follows:
L=D-W
wherein n is the number of users, and the calculation formulas of W and D are respectively:
2.4, solving the following formula to obtain an output weight value beta:
min||β|| 2 +λTr(β T H T LHβ),s.t.(Hβ) T Hβ=I
wherein λ represents a compromise parameter, tr (-) represents the trajectory of the matrix, and to avoid the degenerate solution of β, the above equation can be replaced by the generalized eigenvalue problem:
(I+λH T LH)v=γH T Hv
wherein I is a unit matrix, v is a characteristic vector, and gamma is a characteristic value; solving the above formula, l eigenvectors can be obtained, and the output weight β will be composed of l generalized eigenvectors v, that is:
2.5 after obtaining the output weight beta, calculating a low-dimensional output matrix E through the following formula, namely mapping the input data to a user load characteristic set of an l-dimensional characteristic space through an US-ELM algorithm;
4. the method for classifying user load characteristics of a power distribution network based on data acquisition according to claim 3, wherein the method comprises the following steps: the step 3 comprises the following steps:
3.1, testing different types of numbers by using DB as a judgment standard in an initial stage, determining the type number C, and setting other flexible parameters of a control algorithm, including an initial membership matrix u 0 A weighting index q;
3.2, calculating a clustering center a:
wherein p is an overlapGeneration number of times, e k Is a data set in E, u ik Represents the ith cluster center and the kth data e k C is a clustering number;
3.3 modifying the membership matrix u:
step 3.4 whenIf so, stopping iteration, otherwise, executing the step 3.2, and performing the next iteration, wherein the iteration times p = p +1 until the errors of the two membership matrixes are within a specified range, and at the moment, clustering the user load characteristic set into a C type;
step 3.5, the membership degrees of the cluster centers of the user load feature sets in each class are ranked, and the user load feature set v with high membership degree is selected i (i =1,2,.., h) and clustering result label t i (I =1,2.,. H) are combined into training samples I for a vector machine (SVM) classifier, namely:
and 3.6, classifying the residual user load feature set with the low membership value by using the trained SVM classifier to obtain a classification label, and replacing the clustering label of the residual user load feature set with the low membership value to obtain more accurate classification of the user load feature set by the improved FCM algorithm.
5. The method for classifying user load characteristics of a power distribution network based on data acquisition according to claim 4, wherein the method comprises the following steps: the averaging process in step 4 is shown in the following formula:
wherein, Y i For the typical load data set after the i-th class user is homogenized, p is the number of users contained in the i-th class, x ij Load data points for the ith user are shown.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211045414.7A CN115481786A (en) | 2022-08-30 | 2022-08-30 | Power distribution network user load characteristic classification method based on data acquisition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211045414.7A CN115481786A (en) | 2022-08-30 | 2022-08-30 | Power distribution network user load characteristic classification method based on data acquisition |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115481786A true CN115481786A (en) | 2022-12-16 |
Family
ID=84423006
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211045414.7A Pending CN115481786A (en) | 2022-08-30 | 2022-08-30 | Power distribution network user load characteristic classification method based on data acquisition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115481786A (en) |
-
2022
- 2022-08-30 CN CN202211045414.7A patent/CN115481786A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111476435B (en) | Charging pile load prediction method based on density peak value | |
CN117034179B (en) | Abnormal electric quantity identification and tracing method and system based on graph neural network | |
CN111461921B (en) | Load modeling typical user database updating method based on machine learning | |
CN108345908A (en) | Sorting technique, sorting device and the storage medium of electric network data | |
CN110738232A (en) | grid voltage out-of-limit cause diagnosis method based on data mining technology | |
CN110766320A (en) | Method and device for evaluating operation safety of airport intelligent power grid | |
CN110705859A (en) | PCA-self-organizing neural network-based method for evaluating running state of medium and low voltage distribution network | |
CN111428766B (en) | Power consumption mode classification method for high-dimensional mass measurement data | |
CN110991689A (en) | Distributed photovoltaic power generation system short-term prediction method based on LSTM-Morlet model | |
CN114880939A (en) | Intelligent prediction method and device for service life of power battery | |
CN109858667A (en) | It is a kind of based on thunder and lightning weather to the short term clustering method of loading effects | |
CN117493922A (en) | Power distribution network household transformer relation identification method based on data driving | |
CN115481786A (en) | Power distribution network user load characteristic classification method based on data acquisition | |
CN116307111A (en) | Reactive load prediction method based on K-means clustering and random forest algorithm | |
CN115374991A (en) | Improved statistics upscale-based photovoltaic cluster output prediction method | |
Wang et al. | Analysis of user’s power consumption behavior based on k-means | |
CN113222339A (en) | Energy consumption difference analysis method and system for enterprises of the same type based on measured panoramic data | |
CN110852628A (en) | Rural medium and long term load prediction method considering development mode influence | |
Li et al. | Multi-source heterogeneous log fusion technology of power information system based on big data and imprecise reasoning theory | |
Fan et al. | Prediction and Analysis of Power User Energy Consumption Based on Demand Side Management | |
Anuar et al. | Determination of typical electricity load profile by using double clustering of fuzzy c-means and hierarchical method | |
CN114881429B (en) | Data-driven-based method and system for quantifying line loss of transformer area | |
CN113283702B (en) | Power flow state evaluation method and device integrating safety and stability characteristics of power system | |
CN117094478B (en) | Energy scheduling management method, device, equipment and storage medium | |
CN117993732A (en) | Ordered electricity utilization decision method based on user electricity utilization behavior analysis |
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 |