CN107730059A - The method of transformer station's electricity trend prediction analysis based on machine learning - Google Patents
The method of transformer station's electricity trend prediction analysis based on machine learning Download PDFInfo
- Publication number
- CN107730059A CN107730059A CN201711222095.1A CN201711222095A CN107730059A CN 107730059 A CN107730059 A CN 107730059A CN 201711222095 A CN201711222095 A CN 201711222095A CN 107730059 A CN107730059 A CN 107730059A
- Authority
- CN
- China
- Prior art keywords
- electricity
- transformer station
- machine learning
- trend prediction
- data
- 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 title claims abstract description 73
- 238000004458 analytical method Methods 0.000 title claims abstract description 32
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000010801 machine learning Methods 0.000 title claims abstract description 15
- 238000012549 training Methods 0.000 claims abstract description 21
- 230000000694 effects Effects 0.000 claims abstract description 9
- 238000000605 extraction Methods 0.000 claims description 10
- 238000006243 chemical reaction Methods 0.000 claims description 9
- 230000002441 reversible effect Effects 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 3
- 238000012512 characterization method Methods 0.000 abstract description 4
- 238000005457 optimization Methods 0.000 abstract description 4
- 230000000052 comparative effect Effects 0.000 abstract description 2
- 238000013461 design Methods 0.000 description 5
- 238000003066 decision tree Methods 0.000 description 4
- 230000001373 regressive effect Effects 0.000 description 4
- 230000004927 fusion Effects 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000003869 coulometry Methods 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000003252 repetitive effect Effects 0.000 description 2
- 238000000714 time series forecasting Methods 0.000 description 2
- 238000012300 Sequence Analysis Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 230000008569 process Effects 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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06F18/2148—Generating training patterns; Bootstrap methods, e.g. bagging or boosting characterised by the process organisation or structure, e.g. boosting cascade
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Water Supply & Treatment (AREA)
- Game Theory and Decision Science (AREA)
- Evolutionary Biology (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Development Economics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Public Health (AREA)
- Entrepreneurship & Innovation (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Medical Informatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of method of transformer station's electricity trend prediction analysis based on machine learning, including:The factor of influence of transformer station's electricity trend prediction is analyzed, it is determined that the characteristic quantity type needed for structure model;Based on the electric quantity data gathered and characteristic quantity type structure various dimensions characteristic quantity data set;Build GBDT and Adaboost integrated predictive models, and the prediction effect of the value comparative assessment model using root-mean-square error.The method of transformer station's electricity trend prediction analysis provided by the invention based on machine learning, taking into full account influences the possible characterization factor of prediction effect so that forecast analysis is more accurate;And GBDT, Adaboost Ensemble Learning Algorithms based on recurrence are used, data over-fitting is prevented, and the sustainable training of forecast model, analysis and optimization can be realized.
Description
Technical field
The present invention relates to electric energy meter metering field, and in particular to a kind of transformer station's electricity trend prediction based on machine learning
The method of analysis.
Background technology
Continuous development and people's living standards continue to improve with national economy, annual power consumption are also increasing steadily
It is long.Analysis to transformer station's electric quantity data is closely related to electric power enterprise and regional economy interests, and academic, industry is to transformer station
Coulometric analysis and prediction have been attempted always.Existing transformer station's coulometric analysis and Forecasting Methodology are mainly included based on classical reason
Analyzed by the prediction algorithm analysis of formula and based on univariate time series forecasting:
1st, the prediction algorithm analysis Main Basiss historical data mathematical analysis based on classical theory formula and combination business point
Analysis obtains a set of classical theory formula, and transformer station's power quantity predicting is carried out based on the formula.
This kind of method the problem of it is obvious, be exactly this algorithm based on classical theory formula, it is difficult to reflection in the future
Cause more X factors of data movement.It only considered historical data, it is difficult to which the real data in application future is iterated
And amendment, therefore there is certain limitation.
2nd, based on the analysis of univariate time series forecasting mainly by this single data based on electricity and time the two
Dimension uses typical time sequence analysis algorithm such as Three-exponential Smoothing algorithm and difference autoregressive moving average algorithm.
This kind of method can solve the problem that iteration and the amendment that model parameter is carried out using following real data, can merge not
Carry out prediction of the wobble variation rule realization of real data to data, but this method only considers this single factors of electricity, does not examine
Considering causes the other factors of electric quantity change, such as festivals or holidays, temperature, therefore also has certain limitation.
The content of the invention
In view of the shortcomings of the prior art, it is a primary object of the present invention to:Sum is designed based on multidimensional data Feature Engineering
According to structure, a kind of method of reliable transformer station power quantity predicting is formed using the mode of machine learning.
To realize object defined above, the invention discloses a kind of transformer station's electricity trend prediction analysis based on machine learning
Method, comprise the following steps:
S1. the factor of influence of transformer station's electricity trend prediction is analyzed, it is determined that the characteristic quantity type needed for structure model, described
Characteristic quantity type includes festivals or holidays electricity, month electricity, all electricity, day electricity, meteorological data, model lag period;
S2. based on the electric quantity data gathered and characteristic quantity type structure various dimensions characteristic quantity data set, wherein, institute
State average, variance, maximum, minimum that festivals or holidays electricity, month electricity, all electricity, day electricity are based on respective timing node
Feature extraction is carried out in the characteristic of value;
S3. generation model assesses data set, and the data set includes training set and test set;
S4. the GBDT Ensemble Learning Algorithms structure GBDT forecast models based on recurrence, are iterated using the training set
Training obtains final strong learner;
S5. based on Adaboost Ensemble Learning Algorithms structure Adaboost forecast models, changed using the training set
Generation training obtains final strong classifier;
S6. the power quantity predicting result of the GBDT forecast models and Adaboost forecast models is obtained, and according to the survey
Examination collection, the root-mean-square error of the GBDT forecast models and Adaboost forecast models is calculated respectively;
S7. the prediction effect of the value assessment prediction model of the root-mean-square error is used, the root-mean-square error is smaller, its
The prediction result of forecast model is more accurate.
Preferably, the electric quantity data includes transformer station ID, voltage class, date, electricity.
Preferably, the characteristic that the step S2 is also included in the skewness and kurtosis based on month of month electricity is entered
Row feature extraction.
Preferably, the meteorological data includes temperature and humidity.
Preferably, the electricity includes the positive electricity after positive electricity, reverse electricity, conversion, the reverse electricity after conversion
Amount.
Preferably, the step S2 is also included to carrying out feature extraction based on the temperature of certain day, the characteristic of humidity.
Compared with prior art, the beneficial effects of the present invention are on the basis of mass data, take into full account to influence in advance
Survey the possible characterization factor of effect, and realize for these data sets feature carry out engineering design and data structure and it is more
Dimension data merges, and is finally carried out using GBDT, Adaboost Ensemble Learning Algorithms structure integrated predictive model based on recurrence pre-
Survey, data over-fitting can be prevented, and the sustainable training of forecast model, analysis and optimization can be realized, lift predictive ability.
Brief description of the drawings
Fig. 1 is part sample data exemplary plot disclosed in an exemplary embodiments of the invention;
Fig. 2 is partial data collection exemplary plot disclosed in an exemplary embodiments of the invention.
Embodiment
In view of deficiency of the prior art, inventor is able to propose the present invention's through studying for a long period of time and largely putting into practice
Technical scheme.The technical scheme, its implementation process and principle etc. will be further explained as follows.
The present invention utilizes historical data and machine learning mode based on the design of multidimensional data Feature Engineering and data structure
Realize the prediction to transformer station's future electricity.Including:
1. the design of multidimensional data Feature Engineering and data structure, more dimensions are carried out based on the transformer station's electric quantity data gathered
According to fusion;
The characterization factor that prediction effect may be influenceed is analyzed, for example, festivals or holidays electricity, month electricity, all electricity,
Day electricity, temperature, humidity etc., break through tradition, the limitation of single features.
2. GBDT, Adaboost Ensemble Learning Algorithms based on recurrence are analyzed come implementation model, train and assessed.
Regressive prediction model mould of the GBDT and Adaboost Ensemble Learning Algorithms structure based on time series based on recurrence
Type, data over-fitting is prevented, and the sustainable training of algorithm model, analysis and optimization can be realized.
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that specific embodiment described herein is only to explain the present invention, and
It is not used in the restriction present invention.
The data of transformer station's electricity trend prediction analysis method mostly come from national grid disclosed in the embodiment of the present invention
The electric quantity data based on transformer station's level of electric energy meter measurement monitoring system, and in particular to data type includes transformer station ID, electricity
After pressing grade, date, electricity, electricity to specifically include positive electricity, reverse electricity, the positive electricity after handling and changing, conversion
Reverse electricity, part sample data example is as shown in Figure 1.The prediction analysis method specifically includes:
First, the multidimensional data characteristic Design based on transformer station's electric quantity data
By the daily positive electricity of different substation of collection and reverse electric quantity data, reality is converted to by business processing
Border prediction data, that is, the reverse electricity after positive electricity, conversion after changing.Based on these data, analysis transformer station electricity becomes
The factor of influence of gesture prediction simultaneously builds the characteristic quantity type of regressive prediction model needs, including:
Festivals or holidays electricity, carried out by important festivals time point based on initial data in average, variance, maximum, minimum
In value characteristic extraction, such as the Spring Festival, National Day, the Mid-autumn Festival, International Labour Day, the Ching Ming Festival period electric quantity data;
Month electricity, belong to certain month average of the time point initial data based on month, variance, maximum, minimum value,
Characteristic extraction in the degree of bias, kurtosis;
All electricity, belong to a certain all time point initial data based in all averages, variance, maximum, minimum value
Characteristic extraction;
Day electricity, belongs to the time point initial data of one day based in daily average, variance, maximum, minimum value
Characteristic is extracted;
Meteorological data, based on the temperature of certain day, Humidity Features data fusion and extraction;
The model lag period is 7 days.
By taking the positive electricity structure regressive prediction model after conversion as an example, the characteristic quantity partial data collection example extracted is such as
Shown in Fig. 2.
2nd, transformer station's power quantity predicting analysis and assessment based on regression model
Characteristic quantity data are extracted based on above-mentioned steps, using GBDT, AdaBoost Ensemble Learning Algorithms mould based on recurrence
Type come build for the positive electricity after conversion, conversion after reverse electricity the regressive prediction model based on time series, tool
Body includes:
1. build the GBDT forecast models based on recurrence
GBDT (Gradient Boosting Decison Tree) full name is gradient lifting decision tree, and so-called gradient carries
Liter is the multiple weak learners of continuous grey iterative generation and integrates to Distribution Algorithm to obtain the calculation of the strong learner of final one by preceding
Method, belong to the category of integrated study, while the base learner that GBDT is used is decision tree.Decision Tree algorithms are compared to others
Algorithm needs less Feature Engineering, for example, can it goes without doing feature normalization, can processing field lacks well data, no
Whether interdepended between care feature.Decision tree learning device and gradient method for improving, which are combined, also more effectively to be avoided
The situation of fitting.
GBDT regression algorithm flow principle analysis is as follows:
Training set sample the T={ (x of input1,y1),(x2,y2),....,(xm,ym), maximum iteration T, lose letter
Number L.
The strong learner of output is f (x)
(1) weak learner is initialized
(2) for iterations t=1,2,3, L, T, following iteration is done:
I) for sample i=1,2 ..., m, negative gradient is calculated to be fitted the approximation of each iteration residual error, the of t wheels
The negative gradient of the loss function of i sample is expressed as:
II (x) is utilizedi,rti) (i=1,2 ..., m), are fitted a Cart regression tree, obtain the t regression tree, its is right
The region for the leaf node answered is Rtj, j=1,2,3, L, J.Wherein J is the number of regression tree t leaf node.
III) for leaf node region j=1,2,3, L, J, best-fit values are calculated
IV strong learner) is updated
3) final strong learner f (x) expression formula is obtained
2. build AdaBoost forecast models
AdaBoost (Adaptive Boosting) full name is adaptive enhancing learner, and it is to train number in initialization
According to weights distribution after, carry out continuous repetitive exercise Weak Classifier, i.e., if some sample point is classified exactly, that
Under construction in a training set, its weights are just lowered;On the contrary, if some sample point is not classified exactly,
So its weights are just improved.Then, the sample set that right value update is crossed be used to train next grader, so repeatedly
Go on until iteration reaches some predetermined sufficiently small error rate or reaches preassigned maximum iteration to generation
Afterwards, final strong classifier will be combined into Distribution Algorithm before multiple Weak Classifiers application before.
AdaBoost algorithm flow principle analysis is as follows:
1) the weights distribution of training data is initialized.
Each training sample is endowed identical weights when most starting:W=1/N, it is total sample number
D1=(w11,w12,w13...w1i...,w1N), w1i=1/N, i=1,2,3...N
D1Represent the weights of each sample of first time iteration, w1iRepresent, the weights of i-th of sample during the 1st iteration.
2) more wheel iteration are carried out, represent iterations, m=1,2,3L, M, M is integer.
I) using with weights distribution DmTraining dataset study, obtain basic classification device, it is as follows:
Gm(x):χ∈{-1,+1}
II G) is calculatedm(x) the error in classification rate on training dataset
From above-mentioned formula, Gm(x) the error rate e on training datasetmIt is exactly by Gm(x) power of misclassification sample
It is worth sum.
III G) is calculatedm(x) coefficient, αmRepresent Gm(x) significance level in final classification device, relational expression represent such as
Under:
From above-mentioned formula, emWhen≤1/2, αm>=0, and αmWith emReduction and increase, it is meant that error in classification
Effect of the smaller basic classification device of rate in final classification device is bigger.
IV the weights for) updating training dataset are distributed for next round iteration so that by basic classification device Gm(x) misclassification
The weights increase of sample, and reduced by the weights of correct classification samples.
Dm+1=(wm+1,1,wm+1,2...wm+1,i...,wm+1,N)
Dm+1The weights of sample, w when being for next iterationm+1,iWhen being next iteration, the weights of i-th of sample.Its
In, yiRepresent classification corresponding to i-th of sample, Gm(xi) represent Weak Classifier to sample xiClassification.ZmIt is standardizing factor,
So that Dm+1As a probability distribution.
3) each Weak Classifier is combined
Obtain final strong classifier
Sign functions are used to ask the positive and negative of numerical value, and numerical value is more than 0, functional value 1;Less than 0, functional value is -1;Equal to 0,
Functional value is 0.
3. the assessment of model
After obtaining integrated predictive model by above repetitive exercise, corresponding electricity trend prediction result can be obtained, is adopted
Calculated with RMSE (root-mean-square error) index as follows come the prediction effect of comparative assessment model, formula:
WhereinRepresent power quantity predicting data, eiRepresent actual electric quantity data, n represents prediction number of days, and root-mean-square error is smaller
Represent that the deviation of corresponding model is smaller, its predictive ability is more accurate.
The method of transformer station's electricity trend prediction analysis disclosed by the invention based on machine learning, in the base of mass data
On plinth, take into full account influence the possible characterization factor of prediction effect, and realize for these data sets Feature Engineering design and
Data are built and multidimensional data fusion;The structure of forecast model is calculated using GBDT, Adaboost integrated study based on recurrence
Method is integrated, and can effectively prevent data over-fitting, and can realize the sustainable training of forecast model, analysis and optimization, so as to
Obtain preferable, reliable prediction result.
It should be appreciated that the technical concepts and features of above-described embodiment only to illustrate the invention, its object is to allow be familiar with this
The personage of item technology can understand present disclosure and implement according to this, and it is not intended to limit the scope of the present invention.It is all
The equivalent change or modification made according to spirit of the invention, it should all be included within the scope of the present invention.
Claims (6)
1. the method for transformer station's electricity trend prediction analysis based on machine learning, it is characterised in that comprise the following steps:
S1. the factor of influence of transformer station's electricity trend prediction is analyzed, it is determined that the characteristic quantity type needed for structure model, the feature
Measuring type includes festivals or holidays electricity, month electricity, all electricity, day electricity, meteorological data, model lag period;
S2. based on the electric quantity data gathered and characteristic quantity type structure various dimensions characteristic quantity data set, wherein, the section
Holiday electricity, month electricity, all electricity, day electricity are based on the average of respective timing node, variance, maximum, minimum value
Feature extraction is carried out in characteristic;
S3. generation model assesses data set, and the data set includes training set and test set;
S4. the GBDT Ensemble Learning Algorithms structure GBDT forecast models based on recurrence, training is iterated using the training set
Obtain final strong learner;
S5. based on Adaboost Ensemble Learning Algorithms structure Adaboost forecast models, instruction is iterated using the training set
Practice and obtain final strong classifier;
S6. the power quantity predicting result of the GBDT forecast models and Adaboost forecast models is obtained, and according to the test set,
The root-mean-square error of the GBDT forecast models and Adaboost forecast models is calculated respectively;
S7. the prediction effect of the value assessment prediction model of the root-mean-square error is used, the root-mean-square error is smaller, corresponding pre-
The prediction result for surveying model is more accurate.
2. the method for transformer station's electricity trend prediction analysis according to claim 1 based on machine learning, its feature exist
In:The electric quantity data includes transformer station ID, voltage class, date, electricity.
3. the method for transformer station's electricity trend prediction analysis according to claim 1 based on machine learning, its feature exist
In:The characteristic that the step S2 is also included in the skewness and kurtosis based on month of month electricity carries out feature extraction.
4. the method for transformer station's electricity trend prediction analysis according to claim 1 based on machine learning, its feature exist
In:The meteorological data includes temperature and humidity.
5. the method for transformer station's electricity trend prediction analysis according to claim 2 based on machine learning, its feature exist
In:The electricity includes the positive electricity after positive electricity, reverse electricity, conversion, the reverse electricity after conversion.
6. the method for transformer station's electricity trend prediction analysis according to claim 4 based on machine learning, its feature exist
In:The step S2 is also included to carrying out feature extraction based on the temperature of certain day, the characteristic of humidity.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711222095.1A CN107730059A (en) | 2017-11-29 | 2017-11-29 | The method of transformer station's electricity trend prediction analysis based on machine learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711222095.1A CN107730059A (en) | 2017-11-29 | 2017-11-29 | The method of transformer station's electricity trend prediction analysis based on machine learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107730059A true CN107730059A (en) | 2018-02-23 |
Family
ID=61219901
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711222095.1A Pending CN107730059A (en) | 2017-11-29 | 2017-11-29 | The method of transformer station's electricity trend prediction analysis based on machine learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107730059A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108876019A (en) * | 2018-05-31 | 2018-11-23 | 中国电力科学研究院有限公司 | A kind of electro-load forecast method and system based on big data |
CN109727155A (en) * | 2018-08-02 | 2019-05-07 | 平安科技(深圳)有限公司 | Power consumption control method, apparatus, equipment and storage medium based on power quantity predicting |
CN110175637A (en) * | 2019-05-09 | 2019-08-27 | 北京工商大学 | Non-stationary time-series data depth prediction technique, system, storage medium and equipment |
CN110212520A (en) * | 2019-05-24 | 2019-09-06 | 国网天津市电力公司 | A kind of power predicating method based on convolutional neural networks |
CN111242163A (en) * | 2019-12-28 | 2020-06-05 | 苏州浪潮智能科技有限公司 | Method, system and equipment for predicting performance of storage equipment |
CN112183849A (en) * | 2020-09-25 | 2021-01-05 | 贵州乌江水电开发有限责任公司 | Short-term electric power quantity prediction method based on artificial neural network |
CN112308146A (en) * | 2020-11-02 | 2021-02-02 | 国网福建省电力有限公司 | Distribution transformer fault identification method based on operation characteristics |
CN112330024A (en) * | 2020-11-06 | 2021-02-05 | 国网辽宁省电力有限公司 | Electric quantity prediction method based on non-electric quantity and multi-dimensional scene |
US10928797B2 (en) | 2018-11-29 | 2021-02-23 | Industrial Technology Research Institute | Energy consumption management system and energy consumption management method |
CN115358347A (en) * | 2022-09-30 | 2022-11-18 | 山西虚拟现实产业技术研究院有限公司 | Method for predicting remaining life of intelligent electric meter under different subsystems |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102270309A (en) * | 2011-07-27 | 2011-12-07 | 华北电力大学 | Short-term electric load prediction method based on ensemble learning |
CN102270279A (en) * | 2011-07-27 | 2011-12-07 | 华北电力大学 | Short-term power load predicting method |
CN103400203A (en) * | 2013-07-19 | 2013-11-20 | 西安交通大学 | Electric vehicle charging station load prediction method based on support vector machine |
-
2017
- 2017-11-29 CN CN201711222095.1A patent/CN107730059A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102270309A (en) * | 2011-07-27 | 2011-12-07 | 华北电力大学 | Short-term electric load prediction method based on ensemble learning |
CN102270279A (en) * | 2011-07-27 | 2011-12-07 | 华北电力大学 | Short-term power load predicting method |
CN103400203A (en) * | 2013-07-19 | 2013-11-20 | 西安交通大学 | Electric vehicle charging station load prediction method based on support vector machine |
Non-Patent Citations (2)
Title |
---|
中国环境监测总站 著: "《环境空气质量预报预警方法技术指南 第2版》", 30 June 2017 * |
谢晶晶: "基于层次模型的电能表管理与数据分析方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108876019A (en) * | 2018-05-31 | 2018-11-23 | 中国电力科学研究院有限公司 | A kind of electro-load forecast method and system based on big data |
CN109727155A (en) * | 2018-08-02 | 2019-05-07 | 平安科技(深圳)有限公司 | Power consumption control method, apparatus, equipment and storage medium based on power quantity predicting |
US10928797B2 (en) | 2018-11-29 | 2021-02-23 | Industrial Technology Research Institute | Energy consumption management system and energy consumption management method |
CN110175637A (en) * | 2019-05-09 | 2019-08-27 | 北京工商大学 | Non-stationary time-series data depth prediction technique, system, storage medium and equipment |
CN110212520A (en) * | 2019-05-24 | 2019-09-06 | 国网天津市电力公司 | A kind of power predicating method based on convolutional neural networks |
CN111242163B (en) * | 2019-12-28 | 2022-05-31 | 苏州浪潮智能科技有限公司 | Method, system and equipment for predicting performance of storage equipment |
CN111242163A (en) * | 2019-12-28 | 2020-06-05 | 苏州浪潮智能科技有限公司 | Method, system and equipment for predicting performance of storage equipment |
CN112183849A (en) * | 2020-09-25 | 2021-01-05 | 贵州乌江水电开发有限责任公司 | Short-term electric power quantity prediction method based on artificial neural network |
CN112308146A (en) * | 2020-11-02 | 2021-02-02 | 国网福建省电力有限公司 | Distribution transformer fault identification method based on operation characteristics |
CN112330024A (en) * | 2020-11-06 | 2021-02-05 | 国网辽宁省电力有限公司 | Electric quantity prediction method based on non-electric quantity and multi-dimensional scene |
CN112330024B (en) * | 2020-11-06 | 2023-09-12 | 国网辽宁省电力有限公司 | Electric quantity prediction method based on non-electric quantity and multi-dimensional scene |
CN115358347A (en) * | 2022-09-30 | 2022-11-18 | 山西虚拟现实产业技术研究院有限公司 | Method for predicting remaining life of intelligent electric meter under different subsystems |
CN115358347B (en) * | 2022-09-30 | 2023-01-31 | 山西虚拟现实产业技术研究院有限公司 | Method for predicting remaining life of intelligent electric meter under different subsystems |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107730059A (en) | The method of transformer station's electricity trend prediction analysis based on machine learning | |
CN112508275B (en) | Power distribution network line load prediction method and equipment based on clustering and trend indexes | |
Wu et al. | Evolving RBF neural networks for rainfall prediction using hybrid particle swarm optimization and genetic algorithm | |
CN109754113A (en) | Load forecasting method based on dynamic time warping Yu length time memory | |
CN108280551A (en) | A kind of photovoltaic power generation power prediction method using shot and long term memory network | |
CN108985965A (en) | A kind of photovoltaic power interval prediction method of combination neural network and parameter Estimation | |
CN106951611A (en) | A kind of severe cold area energy-saving design in construction optimization method based on user's behavior | |
CN109063911A (en) | A kind of Load aggregation body regrouping prediction method based on gating cycle unit networks | |
CN106485262A (en) | A kind of bus load Forecasting Methodology | |
CN109325624A (en) | A kind of monthly electric power demand forecasting method based on deep learning | |
CN110674999A (en) | Cell load prediction method based on improved clustering and long-short term memory deep learning | |
US20210326696A1 (en) | Method and apparatus for forecasting power demand | |
CN108549960A (en) | A kind of 24 hours Methods of electric load forecasting | |
CN106022954A (en) | Multiple BP neural network load prediction method based on grey correlation degree | |
CN113537469B (en) | Urban water demand prediction method based on LSTM network and Attention mechanism | |
CN112149890A (en) | Comprehensive energy load prediction method and system based on user energy label | |
CN109034500A (en) | A kind of mid-term electric load forecasting method of multiple timings collaboration | |
CN109636063A (en) | A kind of method of short-term load forecasting | |
CN111428766B (en) | Power consumption mode classification method for high-dimensional mass measurement data | |
CN115186803A (en) | Data center computing power load demand combination prediction method and system considering PUE | |
CN109754122A (en) | A kind of Numerical Predicting Method of the BP neural network based on random forest feature extraction | |
CN109657846A (en) | Power grid alternative subsidy scale impact factor screening technique | |
CN108122173A (en) | A kind of conglomerate load forecasting method based on depth belief network | |
Liu et al. | Using the ART-MMAP neural network to model and predict urban growth: a spatiotemporal data mining approach | |
CN116187835A (en) | Data-driven-based method and system for estimating theoretical line loss interval of transformer area |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180223 |