CN117034046A - Flexible load adjustable potential evaluation method based on ISODATA clustering - Google Patents
Flexible load adjustable potential evaluation method based on ISODATA clustering Download PDFInfo
- Publication number
- CN117034046A CN117034046A CN202310750284.5A CN202310750284A CN117034046A CN 117034046 A CN117034046 A CN 117034046A CN 202310750284 A CN202310750284 A CN 202310750284A CN 117034046 A CN117034046 A CN 117034046A
- Authority
- CN
- China
- Prior art keywords
- load
- cluster
- user
- clustering
- adjustable potential
- 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 25
- 238000000034 method Methods 0.000 claims abstract description 50
- 230000005611 electricity Effects 0.000 claims abstract description 47
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 38
- 238000004364 calculation method Methods 0.000 claims abstract description 25
- 238000004458 analytical method Methods 0.000 claims abstract description 14
- 238000013210 evaluation model Methods 0.000 claims abstract description 5
- 230000006399 behavior Effects 0.000 claims description 14
- 238000004590 computer program Methods 0.000 claims description 11
- 230000001105 regulatory effect Effects 0.000 claims description 10
- 239000006185 dispersion Substances 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 7
- 238000003860 storage Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000007621 cluster analysis Methods 0.000 claims description 5
- 230000002776 aggregation Effects 0.000 claims description 4
- 238000004220 aggregation Methods 0.000 claims description 4
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 238000012804 iterative process Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000013139 quantization Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 229910001021 Ferroalloy Inorganic materials 0.000 description 1
- 241000282414 Homo sapiens Species 0.000 description 1
- 238000003723 Smelting Methods 0.000 description 1
- 229910000831 Steel Inorganic materials 0.000 description 1
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 1
- 229910052782 aluminium Inorganic materials 0.000 description 1
- 239000004568 cement Substances 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 229910021389 graphene Inorganic materials 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 239000010703 silicon Substances 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- 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
-
- 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)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a flexible load adjustable potential evaluation method based on ISODATA clustering, which comprises the following steps: identifying a user power consumption mode through a user historical load curve, and carrying out overall adjustable potential evaluation of the user load through analysis and judgment of the switching feasibility of the user power consumption mode; providing a user load adjustable potential evaluation model based on an electricity consumption mode clustering algorithm; constructing a user power consumption mode identification method based on an ISODATA algorithm; evaluating the optimal clustering number by using CH indexes; and establishing a user load adjustable potential quantitative calculation model. The flexible load adjustable potential evaluation method based on ISODATA clustering can perform cluster recognition on the user history load curve to realize a typical power consumption mode, and the user load flexible adjustable potential analysis method is adopted to quantitatively calculate the most probable adjustment quantity and the maximum adjustable quantity of the user load of the given day.
Description
Technical Field
The application relates to the technical field of flexible load prediction, in particular to an ISODATA clustering-based flexible load adjustable potential evaluation method.
Background
The flexible load aggregation regulation and control is generally carried out by taking an electric power user as a basic unit, dividing the electric power into agriculture, residents, major industries and general industries and businesses according to electricity prices, wherein each major category is divided into a plurality of minor categories, for example, the industrial user comprises 13 minor categories such as smelting, equipment manufacturing, chemical industry and the like, and steel, ferroalloy, industrial silicon, electrolytic aluminum, cement and the like are finely divided; building users among general business users include large business complexes, hotels (guesthouses), malls, office buildings, hospitals (schools), and other 5 subclasses. The historical electricity load curve of the user can express the electricity consumption behavior of the user explicitly, the electricity load curve of the user is collected through the advanced measuring system, the influence factors of the electricity consumption behavior of the user and the sensitivity degree of the influence factors are hidden in the load curve, and the historical load curve is an explicit expression result of the user after the decision of the influence factors.
The related research relates to a big data clustering algorithm, user electricity utilization mode identification, an adjustable potential quantitative evaluation model and the like. For the application of a big data clustering algorithm for classifying the power utilization behaviors of users, firstly, carrying out dimension reduction on massive big data, and improving the data processing efficiency, wherein dimension reduction means comprise methods of principal component analysis, singular value decomposition and the like; then, a proper clustering algorithm is selected, and the document combines a K-means clustering algorithm, a Back Propagation (BP) neural network, a self-organizing map neural network and a deep trust network method, so that the data processing efficiency can be greatly improved, and the recognition precision of the power utilization mode of a user can be improved. Finally, the optimal cluster number is determined and a proper cluster evaluation index is selected, so that the accuracy of the result is improved.
The user electricity consumption big data contains rich intrinsic rules and derivative information of user electricity consumption behaviors, and has the characteristics of large volume, multiple structure types, high speed, low value density and the like, the data-driven electricity consumption behavior analysis research mode is built by taking the data as resources, and the user electricity consumption behavior rules are extracted to identify the typical electricity consumption mode of the user by combining with high-dimensional statistics and other data analysis methods, so that the method is an efficient and feasible method for evaluating the adjustable potential of the user based on the feasibility of the change of the typical electricity consumption mode of the user.
Disclosure of Invention
This section is intended to outline some aspects of embodiments of the application and to briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section as well as in the description of the application and in the title of the application, which may not be used to limit the scope of the application.
The present application has been made in view of the above-described problems.
Therefore, the technical problems solved by the application are as follows: how to accurately evaluate the load adjustable potential of a user.
In order to solve the technical problems, the application provides the following technical scheme: a flexible load adjustable potential assessment method based on ISODATA clustering comprises the following steps:
identifying a user power consumption mode through a user history load curve, and carrying out overall adjustable potential evaluation of the user load through analysis and judgment of the switching feasibility of the user power consumption mode;
constructing a user load adjustable potential evaluation model based on an electricity consumption mode clustering algorithm;
constructing a user power consumption mode identification method based on an ISODATA algorithm;
evaluating the optimal clustering number by using CH indexes;
and establishing a user load adjustable potential quantitative calculation model.
As a preferable scheme of the flexible load adjustable potential evaluation method based on ISODATA clustering, the application comprises the following steps: the identification of the user electricity consumption mode specifically comprises the following steps:
by clustering the same or similar electricity load curves of the users, different typical electricity usage patterns of the users are identified: the electricity consumption mode represents a load curve cluster formed under the same electricity consumption behavior, and the formula is as follows:
wherein M is i Is the ith power mode, x i j(1～96) The load value of the ith power consumption mode at a certain moment from 1 to 96, j represents the index of a time sequence, and n is the number of time periods in one day;
when the user receives the ordered power utilization signals, the user reacts to the signals, and the power utilization behavior is changed under the constraint condition to realize power utilization mode switching, so that the load is regulated.
As a preferred scheme of the flexible load adjustable potential evaluation method based on the ISODATA clustering, the user load adjustable potential evaluation model comprises the following steps: the method comprises two stages of load clustering and classification calculation;
the load clustering stage: carrying out abnormal data detection and KNN data compensation on the user history load big data to obtain a user history load big data set after data cleaning; calculating the optimal clustering number by using the CH index, and obtaining a clustering result of a user history load curve based on an ISODATA algorithm;
the classification calculation stage: and giving a regulation day, calculating a regulation day load baseline, judging the power consumption mode category to which the given regulation day load baseline belongs based on the membership degree, and quantitatively calculating the flexible and adjustable potential of the user load.
As a preferable scheme of the flexible load adjustable potential evaluation method based on ISODATA clustering, the application comprises the following steps: the quantitative calculation of the user load flexibility adjustable potential comprises the following steps: the most likely adjustment and the maximum adjustable;
the most probable adjustment amounts are specifically: the most possible realization of the adjustable capacity value of the user load is evaluated, and the specific calculation mode is to regulate the difference value of the base line load of the day adjustment period and the lowest load in the electricity utilization mode to which the base line load belongs;
the maximum adjustable amount is specifically: and evaluating the maximum value of the adjustable capacity which can be realized by the user, wherein the specific calculation mode is to regulate the difference value between the baseline load of the daily regulation period and the average load of the minimum load electricity utilization mode.
As a preferable scheme of the flexible load adjustable potential evaluation method based on ISODATA clustering, the application comprises the following steps: the user power consumption mode identification method based on the ISODATA algorithm specifically comprises the following steps:
the desired final aggregation category number K; an initial cluster number Nc; minimum number of samples θ in each clustered sample N The method comprises the steps of carrying out a first treatment on the surface of the Threshold value θ of sample standard deviation in each cluster S If the number is larger than the number, the clustering sample needs to be split; threshold value theta of distance between clustering centers C If the number is smaller than the number, combining the two clustering categories; the maximum logarithm L of the categories can be merged in one merging operation; the total number of iterations I;
after setting the above parameters, the ISODATA algorithm can be divided into the following steps:
input N sample data { x } i =1, 2,3, …, N }, arbitrarily select N from the sample data c The initial cluster center samples { z 1 ,z 2 ,z 3 ,...,z Nc ,}；
Calculating the distances from N samples to each cluster center sample, obtaining the sample cluster center with the minimum distance from each sample, and dividing the sample cluster center into corresponding clusters S j Inside going, if d j ＝mind(x,C j ) X is E S j ；
If S j The number of samples in (a) is less than the minimum number of samples in each clustered sample, θ N Then cancel the cluster, let N c ＝N c 1, simultaneously recalculating distances from N samples to each cluster center sample, and carrying out cluster analysis again; if the requirements are met, the center of the cluster is updated on the basis of obtaining the center of the sample cluster with the minimum sample distance, and the correction formula is as follows:
calculating the average distance between all samples and the corresponding cluster center;
discriminating stop, split or merge operations:
if the iteration number I is enough, ending the algorithm, and outputting a clustering result;
if N c < = K/2 and calculate the maximum value σ of the sample distance standard deviation vector in one cluster jmax ＞θ S Then split operation is carried out;
if N c > =2k and calculate the distance D between individual clusters ij ＜θ C Then carrying out merging operation;
if K/2 is less than N c When the iteration number I is an odd number, splitting operation is carried out; when the iteration number I is even, carrying out merging operation;
and repeating the iterative process until the iteration times I are enough, ending the algorithm, and outputting a clustering result.
As a preferable scheme of the flexible load adjustable potential evaluation method based on ISODATA clustering, the application comprises the following steps: the splitting operation specifically comprises the following steps:
performing splitting operation on the clustering samples, wherein a new cluster center formed by splitting each cluster is as follows:
z 1 ＝z i +f actor ×σ jmax
z 2 ＝z i -f actor ×σ jmax
wherein f actor Is an adjustable factor;
the merging operation specifically comprises the following steps:
the cluster samples are combined, and a new cluster center formed by combining each cluster is as follows:
as a preferable scheme of the flexible load adjustable potential evaluation method based on ISODATA clustering, the application comprises the following steps: and evaluating the optimal cluster number by using the CH index, wherein the method specifically comprises the following steps:
the CH index is defined as:
wherein n is the number of training set samples; x is x i The ith sample is the training set; k is the number of given clusters; trB (k) is the trace of the inter-class dispersion matrix; trW (k) is the trace of the intra-class dispersion matrix; x is the average of all sample data; c j Is the center of the j-th cluster; w (w) j,i The dependence relationship between the ith data point and the jth class cluster is obtained; x is X j And the j-th cluster.
As a preferable scheme of the flexible load adjustable potential evaluation method based on ISODATA clustering, the application comprises the following steps: the user load adjustable potential quantitative calculation model specifically comprises the following steps:
based on the classification of the control daily load of the membership degree, judging the electricity consumption mode of the control daily load base line according to the classification result, wherein the cluster of the control daily load base line is as follows:
wherein c is the number of clusters; x is x t Load baseline for the t-th regulatory day; mu (mu) i Is the center of the ith class cluster; zeta type toy it Is x t Membership to the center of the ith class cluster;
let the number of load curves in the class cluster be theta, define and regulate and control the flexible adjustable potential of the time period of the day t to be delta P im (t) can be expressed as:
mu in the middle m The center of the mth cluster of the same cluster not being in the same cluster with the load baseline; ρ is the cluster to which the regulatory daily load classification based on membership; the number of load curves in the class cluster is theta; psi is the class cluster of the largest difference between classes in class clusters, x q Is the load curve with the longest Euclidean distance in the same cluster as the load base line.
A computer device comprising a memory storing a computer program and a processor implementing the steps of the method as described above when the processor executes the computer program.
A computer readable storage medium having stored thereon a computer program which when executed by a processor realizes the steps of the method as described above.
The application has the beneficial effects that: the flexible load adjustable potential evaluation method based on ISODATA clustering can be used for carrying out cluster recognition on the user history load curve to obtain a typical power consumption mode, and the method based on unsupervised user load portraits is provided to carry out classification recognition on the user power consumption mode, so that the flexible load adjustable potential analysis is more facilitated. And quantitatively calculating the most probable adjustment quantity and the maximum adjustable quantity of the given heliostat user load by adopting a user load flexibility adjustable potential analysis method.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. Wherein:
FIG. 1 is a general flow chart of a flexible load adjustable potential evaluation method based on ISODATA clustering provided by the application;
fig. 2 is a flowchart of an ISODATA algorithm of the flexible load adjustable potential evaluation method based on ISODATA clustering.
Detailed Description
So that the manner in which the above recited objects, features and advantages of the present application can be understood in detail, a more particular description of the application, briefly summarized above, may be had by reference to the embodiments, some of which are illustrated in the appended drawings. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, but the present application may be practiced in other ways other than those described herein, and persons skilled in the art will readily appreciate that the present application is not limited to the specific embodiments disclosed below.
Further, reference herein to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic can be included in at least one implementation of the application. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments.
While the embodiments of the present application have been illustrated and described in detail in the drawings, the cross-sectional view of the device structure is not to scale in the general sense for ease of illustration, and the drawings are merely exemplary and should not be construed as limiting the scope of the application. In addition, the three-dimensional dimensions of length, width and depth should be included in actual fabrication.
Also in the description of the present application, it should be noted that the orientation or positional relationship indicated by the terms "upper, lower, inner and outer", etc. are based on the orientation or positional relationship shown in the drawings, are merely for convenience of describing the present application and simplifying the description, and do not indicate or imply that the apparatus or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present application. Furthermore, the terms "first, second, or third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
The terms "mounted, connected, and coupled" should be construed broadly in this disclosure unless otherwise specifically indicated and defined, such as: can be fixed connection, detachable connection or integral connection; it may also be a mechanical connection, an electrical connection, or a direct connection, or may be indirectly connected through an intermediate medium, or may be a communication between two elements. The specific meaning of the above terms in the present application will be understood in specific cases by those of ordinary skill in the art.
Example 1
Referring to fig. 1, for one embodiment of the present application, there is provided a flexible load adjustable potential evaluation method based on ISODATA clustering, including:
s1: the user power consumption mode is identified through the user history load curve, and the overall adjustable potential evaluation of the user load can be carried out through the user power consumption mode switching feasibility analysis and judgment;
the historical electricity load curve of the user can express the electricity consumption behavior of the user explicitly, the electricity load curve of the user is collected through the advanced measuring system, the influence factors of the electricity consumption behavior of the user and the sensitivity degree of the influence factors are hidden in the load curve, and the historical load curve is an explicit expression result of the user after the decision of the influence factors. By clustering the user's same or similar electricity load curves, different typical electricity usage patterns of the user can be identified, as shown in the following equation.
The user power usage pattern is defined as follows: the electricity consumption mode represents a load curve cluster formed under the same or similar electricity consumption behaviors, and the formula is as follows:
wherein Mi is the ith power utilization mode, and xi.jk is the load value at the kth moment in the ith power utilization mode;
when receiving power system adjusting signals such as orderly power utilization notification, power price excitation and the like, a user reacts to the signals, and changes power utilization behaviors under constraint conditions to realize power utilization mode switching and realize adjustment of loads.
Therefore, the user power consumption mode is identified through the user history load curve, and the overall adjustable potential evaluation of the user load can be performed through power consumption mode switching feasibility analysis and judgment.
S2: providing a user load adjustable potential evaluation model based on an electricity consumption mode clustering algorithm;
the method mainly comprises two stages of load clustering and classification calculation;
in the load clustering stage, firstly, carrying out abnormal data detection and KNN data compensation on the user historical load big data to obtain a user historical load big data set after data cleaning; and further calculating the optimal clustering number by using a Calinski-Harabasz (CH) index, and obtaining a clustering result of a user history load curve based on an ISODATA algorithm, wherein the clustering result comprises a cluster-like number and a clustering center, so as to complete the typical power consumption mode identification of the user.
In the classified calculation stage, a regulation day is given, a regulation day load baseline is calculated, the load baseline is calculated by a load average value of five days before a response day, the class of electricity consumption modes to which the given regulation day load baseline belongs is judged based on membership, and further quantitative calculation of the flexible adjustable potential of the user load is carried out: including the most likely adjustment, the most adjustable.
Further, the most probable adjustment amount is the most probable adjustable capacity value for evaluating the user load, and the specific calculation mode is to regulate the difference value between the baseline load of the day adjustment period and the lowest load in the power consumption mode to which the baseline load belongs; the maximum adjustable quantity is the maximum value of adjustable capacity which can be achieved by the evaluation user, and the specific calculation mode is the difference value between the base line load and the average load of the minimum load electricity utilization mode in the adjustment time period of the adjustment day.
S3: constructing a user power consumption mode identification method based on an ISODATA algorithm;
iterative self-organizing analysis algorithms, also known as ISODATA algorithms, are typical of unsupervised learning. The ISODATA algorithm is a soft clustering method developed from statistical pattern recognition, and the algorithm characteristic is represented by the characteristic that an analysis object searches for essential common characteristics among samples under the condition of initial characteristic blurring. Therefore, the ISODATA algorithm is similar to the process of learning and knowing new things by human beings, the output result is continuously corrected in the process of machine learning, the visual clustering effect is achieved, and finally the expected and reasonable prediction effect can be obtained according to the classification requirement.
The concept of class in cluster analysis should be said to be a comparatively abstract term which represents a collection of elements with similar features. The ISODATA clustering algorithm is a dynamic clustering process, and the overall route is as follows: the method comprises the steps of firstly, carrying out rough clustering on sample objects, and secondly, gradually carrying out classification and merging adjustment on clustering results according to the set requirement of the sample objects on the clustering effect so as to achieve the expected classification effect. However, in many cases, we cannot set the parameter balance between the merging and splitting of the clustering algorithm directly and accurately, and the adjustable parameters are relatively large, and different collocations will produce different results, so that the ISODATA clustering cannot obtain good effects before parameter debugging, and the algorithm has the disadvantage that the process of debugging parameter changes is very complex. It is noted that when the debugging is completed, the clustering result obtained by the method is more accurate than that obtained by the traditional clustering method. The reason for this is that the algorithm process integrates the constraint conditions set by us, so that the clustering effect can be more humanized.
Parameters to be set in the clustering process of the ISODATA clustering algorithm comprise the following contents: the desired final aggregation category number K; an initial cluster number Nc; minimum number of samples θ in each clustered sample N The method comprises the steps of carrying out a first treatment on the surface of the Threshold value θ of sample standard deviation in each cluster S It can be understood that the splitting coefficient, if greater than this number, the clustered samples need to be split; threshold value theta of distance between clustering centers C It can be understood that the merging coefficient, if smaller than this number, merges the two cluster categories; the maximum logarithm L of the categories can be merged in one merging operation; total number of iterations I is performed. After setting the above parameters, the ISODATA algorithm can be divided into the following 7 steps in total.
Step 1: input N sample data { x } i =1, 2,3, …, N }, arbitrarily select N from the sample data c The initial cluster center samples { z 1 ,z 2 ,z 3 ,...,z Nc ,}。
Step 2: calculating the distance between N samples and each cluster center sample (Euclidean distance is calculated here), obtaining the sample cluster center with the minimum sample distance, and dividing the sample cluster center into corresponding clusters S j Inside. If d j ＝mind(x,C j ) X is E S j 。
Step 3: if S j The number of samples in (a) is less than theta N Then cancel the cluster, let N c ＝N c -1, simultaneously jumping to the step 2 above, and re-performing cluster analysis; if the requirements are met, the next step is carried out.
Step 4: updating the center of the cluster on the basis of the step 2, and correcting the formula to be:
step 5: calculating the average distance between all samples and the corresponding cluster center:
step 6: discriminating stop, split or merge operations:
1) If the iteration number I is enough, the algorithm is ended, and a clustering result is output
2) If N c < = K/2), then the process goes to step 7, where the splitting step is started.
3) If N c > =2k, then go to step 8, start the merging step.
4) If K/2 < N c When the iteration number I is an odd number, switching to the step 7 to perform splitting operation; and when the iteration number I is even, switching to the step 8, and carrying out merging operation.
Step 7: if the maximum value sigma of the sample distance standard deviation vector in a cluster is calculated jmax ＞θ S The clustering sample is split, and a new cluster center formed by splitting each cluster is as follows:
z 1 ＝z i +f actor ×σ jmax
z 2 ＝z i -f actor ×σ jmax
wherein f actor Is an adjustable factor, and is usually 0.5.
Step 8: calculating the distance D between each cluster ij If D ij ＜θ C And carrying out merging operation, wherein a new cluster center is as follows:
step 9: and repeating the iterative process until the iteration times I are enough, ending the algorithm, and outputting a clustering result.
S4: evaluating the optimal clustering number by using CH indexes;
the optimal clustering number is critical to the clustering analysis, the clustering result is evaluated based on indexes of a sample geometry of the data set, the optimal clustering number is selected according to the statistical characteristics of the data set and the clustering result, the optimal clustering number is evaluated by using CH indexes, the CH indexes describe compactness through an intra-class dispersion matrix, the bigger CH represents tighter class, the more dispersed class is, namely the better clustering result is obtained, and the CH indexes are defined as:
wherein n is the number of training set samples; k is the ith sample of the training set; k is the number of given clusters; trB (k) is the trace of the inter-class dispersion matrix; trW (k) is the trace of the intra-class dispersion matrix; x is the average of all sample data; c j Is the center of the j-th cluster; w (w) j,i The dependence relationship between the ith data point and the jth class cluster is obtained; x is X j And the j-th cluster.
S5: establishing a user load adjustable potential quantitative calculation model;
1) Classification of regulatory daily loads based on membership
Judging the electricity consumption mode of the regulating daily load baseline according to the classification result, wherein the regulating daily load baseline belongs to the cluster l as follows:
wherein c-the number of clusters (number); xt, which is the load baseline of the t regulation day, is the average value of the load of n working days before the regulation day, and generally takes n=5; mu i is the center of the ith cluster; ζit-the degree of membership of xt to the center of the ith class cluster.
2) Adjustable potential quantization calculation model
Let the cluster to which the load control day belongs be Cm, the center of the cluster be μm, the number of load curves in the cluster be θ, the load curve of the t= [ tstart, tend ] adjustment period be xt, the flexibility adjustable potential of the defined control day t period be Δpim (t), m=i, i+1, i+2, …, k can be expressed as:
mu in m The center of the mth cluster of the same cluster not being in the same cluster with the load baseline; xl is the load curve with the longest Euclidean distance in the same cluster as the load base line; Δpim (t) (m=i, i+1, i+2, …, k) represents the user (k-i) adjustable potential in different power modes, thus yielding a flexible adjustable potential in different power modes for regulating the period of day t.
Further, a quantization index for describing the user load flexibility adjustable potential is defined: the most likely adjustment amount Δpi (t), the maximum adjustable amount Δpk (t). The most probable adjustment quantity delta Pi is the most probable adjustable capacity value for evaluating the load of the user, the specific calculation mode is to adjust the difference value between the base line load of the daily adjustment period and the lowest load in the cluster to which the base line load belongs, and the index is defined as the most probable adjustment quantity because the adjustment mode is switched between the power utilization modes of the most similar user; the maximum adjustable quantity Δpik (t) is the maximum value of the adjustable capacity achievable by the evaluation user.
Furthermore, an unsupervised user load portrait-based method is provided, and the user electricity utilization mode is classified and identified, so that the flexible load regulation potential analysis is facilitated.
It should be noted that, the specific calculation mode is to regulate the difference value between the base line load and the cluster center with the smallest load value in the day regulation period, and the index is defined as the maximum adjustable quantity because the regulation mode maximally changes the power consumption mode of the user.
The computer device may be a server. The computer device includes a processor, a memory, an Input/Output interface (I/O) and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing data cluster data of the power monitoring system. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for communicating with an external terminal through a network connection. The computer program, when executed by a processor, implements a flexible load adjustable potential assessment method based on ISODATA clustering.
Example 2
In order to verify the beneficial effects of the application, scientific demonstration is carried out through economic benefit calculation and simulation experiments.
First, we contemplate two classes of users, residential and commercial, whose historical power load data behaves as follows in different time periods:
by using the ISODATA clustering algorithm, we can perform cluster analysis on the electricity usage patterns of each user type according to the historical power load data. For example, residential users may have two modes, a "daytime peak" and a "nighttime off-peak", while commercial users may have two modes, a "work time peak" and an "off-work time off-peak".
Then, through power mode switching feasibility analysis, we can predict that each user type may switch from one power mode to another under different influencing factors (e.g., power price adjustment, energy saving policy, weather conditions, etc.). For example, in the case of an increase in electricity prices or popularization of energy saving policies, the user may try to avoid using electricity during peak hours.
Finally, we can calculate the load adjustable potential for each user type from this information. For example, the load differential between the residential subscribers between the "daytime peak" and "nighttime off-peak" modes is their most likely adjustment; their maximum load difference between the "daytime peak" mode and the "nighttime off-peak" mode is their maximum adjustable amount.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile memory may include Read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high density embedded nonvolatile memory, resistive random access memory (ReRAM), magnetic random access memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric memory (Ferroelectric Random Access Memory, FRAM), phase change memory (Phase Change Memory, PCM), graphene memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor referred to in the embodiments provided in the present application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, or the like, but is not limited thereto.
It should be noted that the above embodiments are only for illustrating the technical solution of the present application and not for limiting the same, and although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present application may be modified or substituted without departing from the spirit and scope of the technical solution of the present application, which is intended to be covered in the scope of the claims of the present application.
Claims (10)
1. A flexible load adjustable potential evaluation method based on ISODATA clustering is characterized by comprising the following steps:
identifying a user power consumption mode through a user history load curve, and carrying out overall adjustable potential evaluation of the user load through analysis and judgment of the switching feasibility of the user power consumption mode;
constructing a user load adjustable potential evaluation model based on an electricity consumption mode clustering algorithm;
constructing a user power consumption mode identification method based on an ISODATA algorithm;
evaluating the optimal clustering number by using CH indexes;
and establishing a user load adjustable potential quantitative calculation model.
2. The flexible load adjustable potential assessment method based on ISODATA clustering as claimed in claim 1, wherein the flexible load adjustable potential assessment method is characterized by comprising the following steps: the identification of the user electricity consumption mode specifically comprises the following steps:
by clustering the same or similar electricity load curves of the users, different typical electricity usage patterns of the users are identified: the electricity consumption mode represents a load curve cluster formed under the same electricity consumption behavior, and the formula is as follows:
wherein M is i Is the i-th power mode of use,the load value of the ith power consumption mode at a certain moment from 1 to 96, j represents the index of a time sequence, and n is the number of time periods in one day;
when the user receives the ordered power utilization signals, the user reacts to the signals, and the power utilization behavior is changed under the constraint condition to realize power utilization mode switching, so that the load is regulated.
3. The flexible load adjustable potential assessment method based on ISODATA clustering as claimed in claim 2, wherein: the user load adjustable potential assessment model comprises: the method comprises two stages of load clustering and classification calculation;
the load clustering stage: carrying out abnormal data detection and KNN data compensation on the user history load big data to obtain a user history load big data set after data cleaning; calculating the optimal clustering number by using the CH index, and obtaining a clustering result of a user history load curve based on an ISODATA algorithm;
the classification calculation stage: and giving a regulation day, calculating a regulation day load baseline, judging the power consumption mode category to which the given regulation day load baseline belongs based on the membership degree, and quantitatively calculating the flexible and adjustable potential of the user load.
4. The flexible load adjustable potential assessment method based on ISODATA clustering as claimed in claim 3, wherein: the quantitative calculation of the user load flexibility adjustable potential comprises the following steps: the most likely adjustment and the maximum adjustable;
the most probable adjustment amounts are specifically: the most possible realization of the adjustable capacity value of the user load is evaluated, and the specific calculation mode is to regulate the difference value of the base line load of the day adjustment period and the lowest load in the electricity utilization mode to which the base line load belongs;
the maximum adjustable amount is specifically: and evaluating the maximum value of the adjustable capacity which can be realized by the user, wherein the specific calculation mode is to regulate the difference value between the baseline load of the daily regulation period and the average load of the minimum load electricity utilization mode.
5. The flexible load adjustable potential assessment method based on ISODATA clustering as claimed in claim 4, wherein the flexible load adjustable potential assessment method is characterized by comprising the following steps: the user power consumption mode identification method based on the ISODATA algorithm specifically comprises the following steps:
the desired final aggregation category number K; an initial cluster number Nc; minimum number of samples θ in each clustered sample N The method comprises the steps of carrying out a first treatment on the surface of the Threshold value θ of sample standard deviation in each cluster S If the number is larger than the number, the clustering sample needs to be split; threshold value theta of distance between clustering centers C If the number is smaller than the number, combining the two clustering categories; the maximum logarithm L of the categories can be merged in one merging operation; the total number of iterations I;
after setting the above parameters, the ISODATA algorithm can be divided into the following steps:
input N sample data { x } i =1, 2,3, …, N }, arbitrarily select N from the sample data c Initial cluster center sample
Calculating the distance from N samples to each cluster center sample to obtain each sampleThe sample cluster center with the minimum distance is divided into corresponding clusters S j Inside going, if d j ＝mind(x,C j ) X is E S j ；
If S j The number of samples in (a) is less than the minimum number of samples in each clustered sample, θ N Then cancel the cluster, let N c ＝N c 1, simultaneously recalculating distances from N samples to each cluster center sample, and carrying out cluster analysis again; if the requirements are met, the center of the cluster is updated on the basis of obtaining the center of the sample cluster with the minimum sample distance, and the correction formula is as follows:
calculating the average distance between all samples and the corresponding cluster center;
discriminating stop, split or merge operations:
if the iteration number I is enough, ending the algorithm, and outputting a clustering result;
if N c < = K/2 and calculate the maximum value σ of the sample distance standard deviation vector in one cluster jmax ＞θ S Then split operation is carried out;
if N c > =2k and calculate the distance D between individual clusters ij ＜θ C Then carrying out merging operation;
if K/2 is less than N c When the iteration number I is an odd number, splitting operation is carried out; when the iteration number I is even, carrying out merging operation;
and repeating the iterative process until the iteration times I are enough, ending the algorithm, and outputting a clustering result.
6. The flexible load adjustable potential assessment method based on ISODATA clustering as claimed in claim 5, wherein the flexible load adjustable potential assessment method is characterized by comprising the following steps: the splitting operation specifically comprises the following steps:
performing splitting operation on the clustering samples, wherein a new cluster center formed by splitting each cluster is as follows:
z 1 ＝z i +f actor ×σ jmax
z 2 ＝z i -f actor ×σ jmax
wherein f actor Is an adjustable factor;
the merging operation specifically comprises the following steps:
the cluster samples are combined, and a new cluster center formed by combining each cluster is as follows:
7. the flexible load adjustable potential assessment method based on ISODATA clustering as claimed in claim 6, wherein: and evaluating the optimal cluster number by using the CH index, wherein the method specifically comprises the following steps:
the CH index is defined as:
wherein n is the number of training set samples; x is x i The ith sample is the training set; k is the number of given clusters; trB (k) is the trace of the inter-class dispersion matrix; trW (k) is the trace of the intra-class dispersion matrix; x is the average of all sample data; c j Is the center of the j-th cluster; w (w) j,i The dependence relationship between the ith data point and the jth class cluster is obtained; x is X j And the j-th cluster.
8. The flexible load adjustable potential assessment method based on ISODATA clustering as claimed in claim 7, wherein: the user load adjustable potential quantitative calculation model specifically comprises the following steps:
based on the classification of the control daily load of the membership degree, judging the electricity consumption mode of the control daily load base line according to the classification result, wherein the cluster of the control daily load base line is as follows:
wherein c is the number of clusters; x is x t Load baseline for the t-th regulatory day; mu (mu) i Is the center of the ith class cluster; zeta type toy it Is x t Membership to the center of the ith class cluster;
let the number of load curves in the class cluster be theta, define and regulate and control the flexible adjustable potential of the time period of the day t to be delta P im (t) can be expressed as:
mu in the middle m The center of the mth cluster of the same cluster not being in the same cluster with the load baseline; ρ is the cluster to which the regulatory daily load classification based on membership; the number of load curves in the class cluster is theta; psi is the class cluster of the largest difference between classes in class clusters, x q Is the load curve with the longest Euclidean distance in the same cluster as the load base line.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 8 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310750284.5A CN117034046A (en) | 2023-06-25 | 2023-06-25 | Flexible load adjustable potential evaluation method based on ISODATA clustering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310750284.5A CN117034046A (en) | 2023-06-25 | 2023-06-25 | Flexible load adjustable potential evaluation method based on ISODATA clustering |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117034046A true CN117034046A (en) | 2023-11-10 |
Family
ID=88625125
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310750284.5A Pending CN117034046A (en) | 2023-06-25 | 2023-06-25 | Flexible load adjustable potential evaluation method based on ISODATA clustering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117034046A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117435939A (en) * | 2023-12-14 | 2024-01-23 | 广东力宏微电子有限公司 | IGBT health state evaluation method based on big data |
-
2023
- 2023-06-25 CN CN202310750284.5A patent/CN117034046A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117435939A (en) * | 2023-12-14 | 2024-01-23 | 广东力宏微电子有限公司 | IGBT health state evaluation method based on big data |
| CN117435939B (en) * | 2023-12-14 | 2024-03-08 | 广东力宏微电子有限公司 | IGBT health state evaluation method based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106446967A (en) | Novel power system load curve clustering method | |
| CN107230108A (en) | The processing method and processing device of business datum | |
| CN111178611A (en) | Method for predicting daily electric quantity | |
| Wang et al. | Manifold interpolation for large-scale multiobjective optimization via generative adversarial networks | |
| CN107301604A (en) | Multi-model fusion estimation system | |
| CN109492748A (en) | A kind of Mid-long term load forecasting method for establishing model of the electric system based on convolutional neural networks | |
| CN113762387B (en) | Multi-element load prediction method for data center station based on hybrid model prediction | |
| CN112734135A (en) | Power load prediction method, intelligent terminal and computer readable storage medium | |
| CN109544029A (en) | Analysis method, analytical equipment and the terminal of a kind of area's line loss | |
| CN114066071A (en) | Power parameter optimization method based on energy consumption, terminal equipment and storage medium | |
| CN117034046A (en) | Flexible load adjustable potential evaluation method based on ISODATA clustering | |
| CN111428766B (en) | Power consumption mode classification method for high-dimensional mass measurement data | |
| CN115907195A (en) | Photovoltaic power generation power prediction method, system, electronic device and medium | |
| CN112396301A (en) | Power consumer demand response characteristic control method based on energy big data driving | |
| Zhou et al. | Expanding the prediction capacity in long sequence time-series forecasting | |
| CN109146553A (en) | Spot Price forecasting system and its method based on multi-density cluster and multicore SVM | |
| Zhang et al. | Building Energy Consumption Prediction Based on Temporal-Aware Attention and Energy Consumption States | |
| CN114971092B (en) | Method and device for predicting concrete consumption of main control building of transformer substation | |
| Wei et al. | An instance based multi-source transfer learning strategy for building’s short-term electricity loads prediction under sparse data scenarios | |
| CN113762591B (en) | Short-term electric quantity prediction method and system based on GRU and multi-core SVM countermeasure learning | |
| CN116011655A (en) | Load ultra-short-term prediction method and system based on two-stage intelligent feature engineering | |
| CN115330201A (en) | Power grid digital project pareto optimization method and system | |
| CN116266700A (en) | Prediction method for adjustable potential of industrial load demand response | |
| CN114819454A (en) | Electricity larceny detection method, device, equipment, storage medium and program product | |
| CN113962440A (en) | DPC and GRU fused photovoltaic prediction method and system |
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 |