CN106296315A - Context aware systems based on user power utilization data - Google Patents
Context aware systems based on user power utilization data Download PDFInfo
- Publication number
- CN106296315A CN106296315A CN201610972613.0A CN201610972613A CN106296315A CN 106296315 A CN106296315 A CN 106296315A CN 201610972613 A CN201610972613 A CN 201610972613A CN 106296315 A CN106296315 A CN 106296315A
- Authority
- CN
- China
- Prior art keywords
- data
- module
- power utilization
- context aware
- user power
- 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
- 230000005611 electricity Effects 0.000 claims abstract description 36
- 230000003542 behavioural effect Effects 0.000 claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000005201 scrubbing Methods 0.000 claims abstract description 11
- 238000013481 data capture Methods 0.000 claims abstract description 10
- 238000013500 data storage Methods 0.000 claims abstract description 9
- 210000000352 storage cell Anatomy 0.000 claims abstract description 9
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000012217 deletion Methods 0.000 claims abstract description 5
- 230000037430 deletion Effects 0.000 claims abstract description 5
- 238000000034 method Methods 0.000 claims description 54
- 238000004422 calculation algorithm Methods 0.000 claims description 47
- 230000008569 process Effects 0.000 claims description 14
- 239000011159 matrix material Substances 0.000 claims description 4
- 230000009467 reduction Effects 0.000 claims description 4
- 239000006185 dispersion Substances 0.000 claims description 3
- 230000003595 spectral effect Effects 0.000 claims description 2
- 238000004140 cleaning Methods 0.000 claims 1
- 230000011218 segmentation Effects 0.000 abstract description 8
- 239000000284 extract Substances 0.000 abstract description 5
- 238000005192 partition Methods 0.000 abstract 1
- 230000000875 corresponding effect Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000007418 data mining Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000007621 cluster analysis Methods 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 206010008190 Cerebrovascular accident Diseases 0.000 description 1
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 description 1
- 241000221931 Hypomyces rosellus Species 0.000 description 1
- 238000012369 In process control Methods 0.000 description 1
- 208000006011 Stroke Diseases 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000005266 casting Methods 0.000 description 1
- 238000004138 cluster model Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 239000000686 essence Substances 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000010965 in-process control Methods 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 210000003127 knee Anatomy 0.000 description 1
- 210000000629 knee joint Anatomy 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000011112 process operation Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000017105 transposition Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Tourism & Hospitality (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention discloses a kind of context aware system based on user power utilization data, including data capture unit, setting up context aware model and inference engine unit and data storage cell, data capture unit is provided with data acquisition module, data scrubbing module, data preprocessing module and electricity consumption behavioral data cluster module and Cluster Assessment feedback module;Set up context aware model and be provided with data processing module and model building module with inference engine unit;Data acquisition module is provided with sensor;Data scrubbing module is for amendment or deletion error data;User power utilization data are obtained by sensor, carry out data again processing and corresponding partition clustering, then the model dataization carrying out context aware is set up, understand user power utilization behavior, extract custom power pattern, thus form user's client segmentation system of multidimensional, provide effective reference data for power supply administration person works.
Description
Technical field
The present invention relates to technical field of electric power, particularly relate to a kind of context aware system based on user power utilization data.
Background technology
Chinese scholars, enterprise have done many research in terms of context aware, but the still phase of the application in terms of power industry
To blank.Domestic grid company client segmentation has been also carried out numerous studies, the main application passing through new equipment and the optimization of algorithm
Realize new client segmentation.But correlational study the most simply proposes single new client segmentation dimension, and is not based on user behavior
Model Establishment multidimensional taxonomic hierarchies, and the application of new equipment cannot obtain user power utilization information in time, more can do nothing to help electrical network
Company more precisely grasps the business demand of Electricity customers differentiated service, more can not provide individual character for power grid enterprises for client
Change and value-added service provides and supports.
Therefore, it is necessary to design a kind of new context aware systems based on user power utilization data, to solve above-mentioned technology
Problem.
Summary of the invention
For problem present in background technology, it is an object of the invention to build a big Data Analysis Platform, flat at this
Set up a set of user power utilization context aware system on platform, according to business demand, there is the different client segmentation of design targetedly poor
Alienation service plan.
The technical scheme is that and be achieved in that: a kind of context aware systems based on user power utilization data, including
Data capture unit, set up context aware model and inference engine unit, data storage cell, wherein, described data acquisition list
Unit is provided with data acquisition module, data scrubbing module, data preprocessing module and electricity consumption behavioral data cluster module and cluster
Evaluate feedback module;Described context aware model and the inference engine unit set up is provided with data processing module and mould set up by model
Block;Described data acquisition module is provided with sensor, and sensor uses non-invasive method to obtain user power utilization data;Data scrubbing
Module is used for checking that user power utilization data are the most normal, amendment or deletion error data;Described set up context aware model and push away
Reason engine unit includes contextual information model, inference engine, calls control output;Described data storage cell is used for preserving system
All data.
In technique scheme, the acquisition interval time of described user power utilization data is 15min, 30min or 1h.
In technique scheme, described user power utilization data include customer profile data, metering data and payment data.
In technique scheme, described user power utilization data can temporally, geographic area and electric pressure carry out difference
Acquisition.
In technique scheme, described user power utilization data clusters module can use based on the clustering algorithm divided, layer
Secondary clustering algorithm, density-based algorithms, clustering algorithm based on model, fuzzy clustering algorithm, ant clustering algorithm, spectrum
Clustering algorithm, gaussian clustering method.
In technique scheme, described Cluster Assessment feedback module be provided with mean square error, average suitability degree, cluster discrete
Degree, similarity matrix, Dai Wei-Bouldin index, bunch in quadratic sum and bunch between the ratio of difference.
In technique scheme, described data processing module include data edit, the process of shortage of data value and
Data attribute reduction.
In technique scheme, the process of described shortage of data value includes ignoring the field of missing values, deletion has disappearance
The record of value and use average.
Present invention context aware based on user power utilization data system, including data capture unit, sets up context aware mould
Type and inference engine unit and data storage cell, data capture unit is provided with data acquisition module, data scrubbing module, number
Data preprocess module and electricity consumption behavioral data cluster module and Cluster Assessment feedback module;Set up context aware model and reasoning
Engine unit is provided with data processing module and model building module;Data acquisition module is provided with sensor and obtains user power utilization number
According to, then by data scrubbing module, data preprocessing module and electricity consumption behavioral data cluster module and Cluster Assessment feedback mould
Block carries out data and processes and corresponding cluster, and the model dataization then carrying out context aware is set up, and understands user power utilization behavior,
Extract custom power pattern, thus form user's client segmentation system of multidimensional, provide effective reference for power supply administration person works
Data.
Accompanying drawing explanation
Fig. 1 is present invention context aware based on user power utilization data system schematic.
Detailed description of the invention
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Describe, it is clear that described embodiment is only a part of embodiment of the present invention rather than whole embodiments wholely.Based on this
Embodiment in invention, the every other reality that those of ordinary skill in the art are obtained under not making creative work premise
Execute example, broadly fall into the scope of protection of the invention.
As it is shown in figure 1, a kind of context aware systems based on user power utilization data of the present invention, including data acquisition
Unit 1, set up context aware model and inference engine unit 2 and data storage cell 3.
Wherein, data capture unit 1 is provided with data acquisition module 11, data scrubbing module 12, data preprocessing module 13
With data cluster module 14 and Cluster Assessment feedback module 15;Described context aware model of setting up sets with inference engine unit 2
Having data processing module 21 and model building module 22, it includes contextual information model, inference engine, calls control output;Institute
State data storage cell 3 for preserving all data messages of system.
Described data acquisition module 11 is provided with sensor 4, and sensor 4 uses non-invasive method to obtain frequency spectrum data, point
Analysis obtains out user power utilization data.Here, user power utilization data include customer profile data, metering data and payment data, and
The acquisition interval time of user power utilization data is 15min, 30min or 1h.And the initial option of user power utilization behavioral data is permissible
Temporally the condition such as (moon, season, year), geographic area and electric pressure (high, in, low) is carried out.Additionally, data select also with application
Purpose is relevant.Data scrubbing module 12 is the most normal for the user power utilization behavior curve data checking each user, amendment or
Deletion has the data of apparent error (imperfect, have noise or inconsistent).
Data preprocessing module 13 carries out preliminary process to data, concrete, due to industrial nature and the electricity consumption of user
Behavior varies in size, and may there is greatest differences, the most even differ multiple order of magnitude between user power utilization behavioral data value,
The most treated cluster can affect clustering result quality so that cluster result is unreliable.Therefore must logarithm before data are carried out cluster
Standardize according to collection, sample data is restricted in certain limit.So it is not only convenient for the subsequent treatment of data, it is also possible to carry
High convergence rate is to shorten the operation time of cluster.
Data clusters module 14, enters the user power utilization behavior curve after standardization with clustering algorithm under given cluster numbers
Row cluster.Cluster analysis result is vulnerable to the impact of many factors, such as normalized fashion, the cluster result dependence to data set
Property, the stability of algorithm, the algorithm sensitivity etc. to data input sequence.User power utilization behavior curve cluster first has to determine use
Family electricity consumption behavioral trait index, optional suitable clustering algorithm and determine its corresponding parameter.
The main method generally realized with clustering technique can be divided into following a few class:
1) based on the algorithm divided.Basic thought based on the clustering algorithm divided is: given one contains m object
Data set, division methods will build k packet, and each packet just represents a clustering cluster.And each bunch at least include 1 right
As, each object and must only belong to 1 bunch.For given data set, algorithm is first according to given division to be built
Number creates an initial packet, then uses a kind of method of iteration reorientation to change initial packet to change each time
Enter later packet scheme the most previous good.System combines k-means, k-medoids scheduling algorithm and constructs power consumer
The Clustering Model of typical case's electricity consumption behavior model, it is proposed that Cluster Assessment new method.K-medoids based on ant group optimization combines
Conjunction clustering algorithm overcomes k-medoids algorithm and is easily absorbed in the shortcoming of local optimum, improves the accuracy rate of cluster.Pass through example
Demonstrate feasibility and the effectiveness of this algorithm.Utilize k-means cluster to improve electricity consumption behavior prediction and POWER SYSTEM STATE is estimated
The accuracy of meter result.
2) hierarchical clustering algorithm.That hierarchical method can be divided into cohesion according to the isolation of level difference or division.
The method of cohesion is bottom-up decomposition, first using each object as a single group, is then combined with similar group, until
All of group is merged into one (or meeting certain end condition).The method of division is top-down decomposition, first by all of
Object is placed in a group, and in each step of iteration, a group is split into less group, until final each object is at list
In an only group (or meeting certain end condition).The method that hierarchical clustering combines with fuzzy model determines typical case's electricity consumption
Behavior distribution and class of subscriber.Result shows that this model has the ability overcoming the difficulty run in process control and operation.
Use hierarchy clustering method to determine the electricity consumption behavior classification of power consumer, and demonstrate the effectiveness of institute's extracting method.
3) Name-based Routing.Method based on density from major part division methods different, it be not based on various respectively
The distance of sample, but based on density.Its main thought is: as long as the density of close region (object or the number of data point) exceedes
Certain threshold value, continues to cluster.The method both can filter noise data, it is also possible to find arbitrary shape bunch.Combinative analysis
Method, the method is rejected sole user's exception electricity consumption data with density clustering algorithm, is extracted its typical case's electricity consumption behavior model, and lead to
Cross the experimental verification the method effectiveness when carrying out electricity consumption behavior model and extracting and feasibility.Utilize Density Clustering to resident
User carries out preliminary classification, finally determines resident's ladder segmentation electricity and electricity price crosspiece separating structure.
4) algorithm based on model.Method based on model is by optimizing between given data and some mathematical model
Matching.Mainly include statistical method COBWEB, neural net method SOM (self-organization mapping net).
The Fuzzy Self-organizing Neural Network algorithm improved, this algorithm uses windowing to measure electricity consumption behavior curve to be measured and allusion quotation at peak interval of time
The deviation of type electricity consumption behavior curve, and it has been experimentally confirmed the ability of the method Exception Filter and cluster.
5) fuzzy clustering algorithm.Traditional clustering algorithm is a kind of hard plot, and each object to be identified is strictly divided into often
Individual apoplexy due to endogenous wind, division limits is clearly demarcated.But most of object does not actually have the strictest Attribute transposition, it is in form and class
Genus aspect also exists betweenness.Fuzzy theory is utilized to be referred to as fuzzy cluster analysis to the method processing clustering problem.Fuzzy clustering
Analysis is a kind of improvement to tradition hard plot method, and sample belongs to the degree of membership of each classification and have expressed the centre of sample attribute
Property.Obtain typical case's electricity consumption behavior curve by the sorting technique using FCM and probabilistic neural network, carry for consumer and supplier
Supply useful information.
6) other clustering algorithms.
Pertinent literature proposed the new methods such as ant-clustering, spectral clustering, Gaussian clustering are applied to electric power electricity consumption row in recent years
For in the extraction of model, and achieve certain effect.Ant clustering algorithm, (includes that condensing level gathers with other algorithms most in use
Class algorithm, k-means and genetic algorithm for clustering) compare, result display ant algorithm is substantially better than other, works as scale simultaneously
Or clusters number shows more stable when becoming big.Gauss function regression model, this model make use of the excellent of gaussian sum mixed model
Gesture, improves the accuracy of prediction.But need to solve the function of convergence relations problems between response curve and covariance.
In actual application aspect, the most commonly used with k-means and FCM method, they principles are simple, easily realize, during operation
Between shorter and cluster accuracy higher.It should be noted that every kind of clustering method all has different features, using
A kind of clustering algorithm is not had to be always better than other algorithms when electricity behavior model extracts or other aspects apply.Some of which algorithm quilt
Frequently use often because its easy operation or Clustering Effect are preferable.Also need to the difference according to data type in actual applications
Select suitable clustering algorithm to obtain optimal Clustering Effect.Additionally, too many levels is multifactor during cluster result is extracted
Impact, application time should take in, to determine the concrete clustering algorithm of applicable Resolving probiems.
Cluster Assessment feedback module 15, is analyzed the cluster result obtained in abovementioned steps and evaluates.For cluster
Number, the electricity price classification that can relate to according to user or national economy activity trade classification are given, are so easy to analysis and find electricity
Valency classification and category of employment are divided with the corresponding relation of user power utilization behavioral pattern and the electricity consumption behavior model of electricity consumption behavior curve
Cloth, it is possible to determine typical user's electricity consumption behavioral pattern of each client.
Owing to cluster is a kind of unsupervised process, the electricity consumption behavioral data in data set is unlabelled to liking, it is impossible to
Directly obtain useful structural knowledge information.Therefore, evaluate the quality of cluster result and determine that preferable clustering number is a difficulty
Task.Determine that cluster number most common method is to perform clustering algorithm respectively for several times, so under given different cluster numbers
Preferable clustering number is selected afterwards according to predetermined criterion function.Predetermined criterion function is referred to as Cluster Validity evaluation index.When poly-
After the parameter of class number and clustering algorithm is fixing, can assess and verify electricity consumption behavior cluster knot by Cluster Validity evaluation index
Really.The Cluster Validity evaluation index that neither one is single can process any data set or more preferable than other result.
6 suitabilities are currently mainly had to estimate: mean square error (mean square error, MSE), average suitability degree
(mean index adequacy, MIA), cluster dispersion (clustering dispersion indicator, CDI), phase
Seemingly spend matrix (similarity matrix indicator, SMI), Dai Wei-Bouldin index (Davies-Bouldin
Indicator, DBI), bunch in quadratic sum and bunch between ratio (the ratio of within cluster sum of of difference
Squares between cluster variation, WCBCR), but only still cannot determine poly-completely by these indexs
Quality and the number of clusters of class algorithm are the most suitable.Stability indicator and priority metrics evaluation algorithm and select cluster numbers, can
To be used in combination with other indexs.Use " the knee joint of the index such as CDI, SI (scatter index), MIA and cluster numbers curve linear relationship
Point " determine preferable clustering number, but the determination of this " knee point " has certain fuzziness.Cluster stability indicator is easier to as specifically
Application purpose determines that preferable clustering number provides reference.
Suitable cluster numbers is can determine that, it is achieved according to user power utilization behavioral pattern after evaluating by cluster result and feed back
User classifies and obtains the corresponding distribution of user typical case's electricity consumption behavior model.User's classification and user power utilization behavioral pattern extract
Final goal is to support power system Operation Decision, optimizes and runs, and reduces loss, increases economic efficiency.
Set up the data processing module 21 of context aware model and inference engine unit 2 and model building module 22 to above-mentioned
The data of cluster model after processing, and wherein, data processing module 21 includes data edit, the process of shortage of data value
With data attribute reduction, difference is as follows:
(1) data edit
Before Data Mining, need to obtain the base attribute information of user, from billing data, payment data and calling
No. 1000 data collect generation user behavior data, say, that the data clusters for preparing in data, after data have selected
Carrying out Data Mining again, therefore, Data Mining here is to generate user's base attribute information and behavioural information, i.e. data standard
Carry out in the sample data obtained after standby work.Sample data comprises and obtains after initial data is collected statistics
Some new variables arrived.Before exploring data, we also need to the Data Discretization of continuous, from existing many
Individual variable derives useful single variable.
(2) process of shortage of data value
Shortage of data value refers to know in data set, not have to collect or the value of mistake typing.Generally, for
The field that they are affiliated, these values are invalid.Problems is needed to observe missing values situation, it is considered to prediction after casting out
Result whether have large effect.Referring here to the process problem of missing values, the processing method of missing values has following several:
1. ignoring the field of missing values, make this field be not used in modeling, this is primarily adapted for use in containing a large amount of missing values, and
It it not important field.
2. deleting the record with missing values, this is primarily adapted for use in containing a small amount of missing values.
3. average is used.
Default value replaces missing values or the distribution proportion according to existing just data to derive missing values, and this is for containing more
Missing values and important field are more effective.
(3) data attribute reduction
Data in data base the most all correspond to substantial amounts of attribute, but be not each attribute be available, as
Attribute the least to incoherent attribute or relatedness is used among modeling by fruit, it is possible to reduce the performance of Knowledge Discovery process,
Making calculation cost is that the letter number of geometry level increases, or is allowed to fall into chaos, needs for this to carry out attitude layer.
Pass through model building module 22 after carrying out data attribute stipulations again and set up context aware model and inference engine.?
In prediction modeling process, need to find the association attributes of customer consuming behavior.In data base, the data that need to obtain and use
Have: customer profile data, charging and payment data and other data.
Patterned modeling tool, can use modeling tool to be patterned modeling, automatically be given birth to by modeling tool
Become standardized scenario models, simplify modeling engineering.Realize a kind of inference engine being associated with relational database.Imitate from reasoning
Rate is improved and inference engine is further improved by two aspects of Conflict solving.
The index technology using data base to be correlated with improves the speed of sight reasoning, improves by detecting and solving sight conflict
The accuracy of the reasoning results.Scape of understanding for the method that employing theory and example combine models and inference method.Use modeling
Instrument carries out example modeling, and is analyzed the availability of modeling tool;Use the mode of contrast experiment, from context data amount
With two aspects of scenario models complexity, inference engine is estimated.
Present invention context aware based on user power utilization data system, including data capture unit 1, sets up context aware mould
Type and inference engine unit 2 and data storage cell 3, data capture unit 1 is provided with data acquisition module 11, data scrubbing mould
Block 12, data preprocessing module 13 and electricity consumption behavioral data cluster module 14 and Cluster Assessment feedback module 15;Set up sight
Sensor model and inference engine unit 2 are provided with data processing module 21 and model building module 22;Data acquisition module 11 is provided with
Sensor 4 obtains user power utilization data, then is gathered by data scrubbing module 12, data preprocessing module 13 and electricity consumption behavioral data
Generic module 14 and Cluster Assessment feedback module 15 carry out data and process and corresponding cluster, then carry out the model of context aware
Datumization is set up, and understands user power utilization behavior, extracts custom power pattern, thus forms user's client segmentation system of multidimensional,
Effective reference data is provided for power supply administration person works.
The foregoing is only presently preferred embodiments of the present invention, not in order to limit the present invention, all essences in the present invention
Within god and principle, any modification, equivalent substitution and improvement etc. made, should be included within the scope of the present invention.
Claims (8)
1. a context aware system based on user power utilization data, it is characterised in that: include data capture unit, set up situation
Sensor model and inference engine unit, data storage cell, wherein, described data capture unit is provided with data acquisition module, number
According to cleaning module, data preprocessing module and electricity consumption behavioral data cluster module and Cluster Assessment feedback module;Described foundation
Context aware model and inference engine unit are provided with data processing module and model building module;Described data acquisition module is provided with
Sensor, sensor uses non-invasive method to obtain user power utilization data;Data scrubbing module is used for checking user power utilization number
According to the most normal, amendment or deletion error data;Described context aware model of setting up includes contextual information with inference engine unit
Model, inference engine, call control output;Described data storage cell is used for all data of the system that preserves.
Context aware system based on user power utilization data the most according to claim 1, it is characterised in that: described user uses
The acquisition interval time of electricity data is 15min, 30min or 1h.
Context aware system based on user power utilization data the most according to claim 1, it is characterised in that: described user uses
Electricity data include customer profile data, metering data and payment data.
Context aware system based on user power utilization data the most according to claim 3, it is characterised in that: described user uses
Electricity data can temporally, geographic area and electric pressure carry out different acquisitions.
Context aware system based on user power utilization data the most according to claim 1, it is characterised in that: described user uses
Electricity data clusters module can use based on divide clustering algorithm, hierarchical clustering algorithm, density-based algorithms, based on mould
The clustering algorithm of type, fuzzy clustering algorithm, ant clustering algorithm, spectral clustering, gaussian clustering method.
Context aware system based on user power utilization data the most according to claim 1, it is characterised in that: described cluster is commented
Valency feedback module be provided with mean square error, average suitability degree, cluster dispersion, similarity matrix, Dai Wei-Bouldin index, bunch in
Quadratic sum and bunch between the ratio of difference.
Context aware system based on user power utilization data the most according to claim 1, it is characterised in that: at described data
Reason module includes data edit, the process of shortage of data value and data attribute reduction.
Context aware system based on user power utilization data the most according to claim 7, it is characterised in that: described data lack
The process of mistake value includes ignoring the field of missing values, deleting record and the use average having missing values.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610972613.0A CN106296315A (en) | 2016-11-03 | 2016-11-03 | Context aware systems based on user power utilization data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610972613.0A CN106296315A (en) | 2016-11-03 | 2016-11-03 | Context aware systems based on user power utilization data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106296315A true CN106296315A (en) | 2017-01-04 |
Family
ID=57720691
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610972613.0A Pending CN106296315A (en) | 2016-11-03 | 2016-11-03 | Context aware systems based on user power utilization data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106296315A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106980925A (en) * | 2017-03-09 | 2017-07-25 | 上海海能信息科技有限公司 | A kind of regional power grid dispatching method based on big data |
CN107180363A (en) * | 2017-05-15 | 2017-09-19 | 泰康保险集团股份有限公司 | Data capture method and device |
CN108804696A (en) * | 2018-06-15 | 2018-11-13 | 深圳华建电力工程设计有限公司 | The creation method and its system of power consumer electric appliance fingerprint base |
CN109543706A (en) * | 2018-09-19 | 2019-03-29 | 南昌工程学院 | A kind of conflict analysis System and method for based on rough set theory of big data driving |
CN109726740A (en) * | 2018-12-05 | 2019-05-07 | 国网浙江省电力公司湖州供电公司 | A kind of trade power consumption behavior analysis method based on clustering |
CN111784381A (en) * | 2020-06-09 | 2020-10-16 | 北京邮电大学 | Privacy protection and SOM network-based power customer segmentation method and system |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105574642A (en) * | 2015-11-06 | 2016-05-11 | 广东工业大学 | Smart grid big data-based electricity price execution checking method |
-
2016
- 2016-11-03 CN CN201610972613.0A patent/CN106296315A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105574642A (en) * | 2015-11-06 | 2016-05-11 | 广东工业大学 | Smart grid big data-based electricity price execution checking method |
Non-Patent Citations (4)
Title |
---|
张铁峰等: "电力用户负荷模式提取技术及应用综述", 《电网技术》 * |
张铁峰等: "电力用户负荷模式提取技术及应用综述", 《电网技术》, vol. 40, no. 3, 31 March 2016 (2016-03-31), pages 1 - 2 * |
王林: "基于SM/AMI的电力消费行为模式的分析", 《中国优秀硕士学位论文全文数据库经济与管理科学辑》, 15 December 2013 (2013-12-15), pages 2 * |
蒋良孝等: "《贝叶斯网络分类器:算法与应用》", 北京邮电大学出版社, pages: 4 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106980925A (en) * | 2017-03-09 | 2017-07-25 | 上海海能信息科技有限公司 | A kind of regional power grid dispatching method based on big data |
CN106980925B (en) * | 2017-03-09 | 2020-11-17 | 上海海能信息科技有限公司 | Regional power grid dispatching method based on big data |
CN107180363A (en) * | 2017-05-15 | 2017-09-19 | 泰康保险集团股份有限公司 | Data capture method and device |
CN107180363B (en) * | 2017-05-15 | 2021-01-26 | 泰康保险集团股份有限公司 | Data acquisition method and device |
CN108804696A (en) * | 2018-06-15 | 2018-11-13 | 深圳华建电力工程设计有限公司 | The creation method and its system of power consumer electric appliance fingerprint base |
CN109543706A (en) * | 2018-09-19 | 2019-03-29 | 南昌工程学院 | A kind of conflict analysis System and method for based on rough set theory of big data driving |
CN109726740A (en) * | 2018-12-05 | 2019-05-07 | 国网浙江省电力公司湖州供电公司 | A kind of trade power consumption behavior analysis method based on clustering |
CN111784381A (en) * | 2020-06-09 | 2020-10-16 | 北京邮电大学 | Privacy protection and SOM network-based power customer segmentation method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106296315A (en) | Context aware systems based on user power utilization data | |
CN108280479B (en) | Power grid user classification method based on load characteristic index weighted clustering algorithm | |
CN104572449A (en) | Automatic test method based on case library | |
CN103426120A (en) | Medium and low voltage power distribution network comprehensive evaluation method based on reliability | |
CN111505443B (en) | Low-voltage transformer area line-to-user relationship identification method and device and computer equipment | |
CN107944705B (en) | Full-end reliability calculation method for dividing communication communities based on modularity | |
CN114721833A (en) | Intelligent cloud coordination method and device based on platform service type | |
CN103258027A (en) | Context awareness service platform based on intelligent terminal | |
CN104462318A (en) | Identity recognition method and device of identical names in multiple networks | |
CN108984830A (en) | A kind of building efficiency evaluation method and device based on FUZZY NETWORK analysis | |
García Valverde et al. | Water demand estimation and outlier detection from smart meter data using classification and Big Data methods | |
CN107729939A (en) | A kind of CIM extended method and device towards newly-increased power network resources | |
CN105205052A (en) | Method and device for mining data | |
Zhou et al. | Smart energy management | |
CN110287237B (en) | Social network structure analysis based community data mining method | |
CN109241190A (en) | Electric power big data mixes computing architecture | |
Yiping et al. | An improved multi-view collaborative fuzzy C-means clustering algorithm and its application in overseas oil and gas exploration | |
CN101651567A (en) | Available risk spread evaluating method | |
Ren et al. | Customer segmentation of bank based on data warehouse and data mining | |
CN109492055B (en) | A method of power grid big data warehouse is established based on CIM model | |
Zhang | Robust spectral detection of global structures in the data by learning a regularization | |
Cheng et al. | Using clustering analysis and association rule technology in cross-marketing | |
Cheong et al. | Review of clustering algorithms for microgrid formation | |
CN115907505A (en) | Power internet operation and maintenance early warning method and system based on knowledge graph | |
CN111768066B (en) | Park electric heating load coupling relation analysis method and device based on fusion characteristics |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170104 |