CN109389238B - Ridge regression-based short-term load prediction method and device - Google Patents
Ridge regression-based short-term load prediction method and device Download PDFInfo
- Publication number
- CN109389238B CN109389238B CN201710690551.9A CN201710690551A CN109389238B CN 109389238 B CN109389238 B CN 109389238B CN 201710690551 A CN201710690551 A CN 201710690551A CN 109389238 B CN109389238 B CN 109389238B
- Authority
- CN
- China
- Prior art keywords
- day
- data
- historical
- meteorological
- prediction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
The invention relates to a short-term load prediction method and a short-term load prediction device based on ridge regression, wherein the method comprises the following steps: determining a training sample by using meteorological data of historical days and corresponding load data; inputting meteorological data of a prediction day into a prediction model, and determining load data of the prediction day, wherein the prediction model is obtained by performing ridge regression training on training samples; according to the technical scheme, a sample obtained by screening historical sample data by using a meteorological similarity coefficient is used as a training sample for training a prediction model, and the prediction model is evaluated by using the sample data after the prediction model is obtained; the training model obtained based on the training of the sample without irrelevant sample interference has higher accuracy; the prediction model is used for predicting the load data of the transformer area in the future time period after being evaluated by the evaluation function, so that the accuracy of the prediction result is further ensured.
Description
Technical Field
The invention relates to the technical field of load prediction of power systems, in particular to a short-term load prediction method and device based on ridge regression.
Background
The load prediction determines a certain prediction model according to the historical load value, and determines the load value at a certain future time under the condition of meeting certain precision. Load prediction is an important component of electric power operation research and power distribution network planning performed by an electric power department, is the root of ensuring safe, effective and economic operation of an electric power system, and is a prerequisite for electric power planning and construction; the reasonability of the layout, investment and operation of the power network is mainly influenced by the accuracy of power load prediction, and the economic, safe and reliable operation of the power system can not be accurately predicted without departing from the power load, so that the technical level of load prediction is improved, the power dispatching management is facilitated, the reasonable arrangement of power grid planning and construction is facilitated, and the economic benefit and the social benefit of the power system are improved.
Load prediction has become an important content for implementing intelligent power modernization management. At present, most of power load prediction algorithms adopt an empirical method, the problem of low prediction accuracy exists, and besides, along with the large concentration of mass power utilization data, the traditional data analysis means is not enough to meet the requirement of mass data mining. Therefore, it is desirable to provide a load prediction method based on a reliable algorithm to realize high-precision prediction of the load of the power system.
Disclosure of Invention
The invention provides a short-term load prediction method and a short-term load prediction device based on ridge regression, and aims to determine training samples without interference factors according to historical meteorological data and corresponding historical load data screened by meteorological similarity indexes; and training a prediction model based on a ridge regression algorithm according to the screened training samples, thereby realizing high-precision prediction of the power system load in a certain time period in the future.
The purpose of the invention is realized by adopting the following technical scheme:
a method of short-term load prediction based on ridge regression, the method comprising:
acquiring meteorological data of historical days and predicted days and load data corresponding to the meteorological data of the historical days;
determining a training sample according to a similarity index value matrix of the meteorological data of the historical day and the forecast day and load data corresponding to the meteorological data of the historical day;
and inputting the meteorological data of the prediction day into a prediction model, and determining the load data of the prediction day, wherein the prediction model is obtained by performing ridge regression training on training samples.
Preferably, the meteorological data includes: temperature, humidity, air pressure, wind speed data.
Preferably, the determining the training sample according to the similarity index value matrix of the meteorological data of the historical day and the forecast day and the load data corresponding to the meteorological data of the historical day includes:
determining a similarity coefficient between the historical day and the predicted day according to the similarity index value matrix of the meteorological data of the historical day and the predicted day;
and selecting the meteorological data of the historical day corresponding to the previous n large similarity coefficients in the similarity coefficients and the load data corresponding to the meteorological data as training samples.
Preferably, the similarity index value matrix R of the meteorological data of the historical day i and the predicted day f is determined by the following formula fi :
R fi =[r i1j ,r i2j ,r i3j ,r i4j ] T
Wherein r is i1j Euclidean distance, r, of the jth meteorological factor for historical day i versus predicted day f i2j The correlation coefficient r of the historical day i to the jth meteorological factor of the predicted day i3j The Euclidean distance r is obtained by first-order difference between the jth meteorological factor of the historical day i and the predicted day and the meteorological factor value of the previous day i4j The correlation coefficient is obtained by first-order difference between the historical day i and the jth meteorological factor of the forecast day and the meteorological factor value of the previous day, i belongs to [1, n ∈],j∈[1,m]N is the total days of the historical days, and m is the total number of meteorological factors in meteorological data;
said r i1j 、r i2j 、r i3j And r i4j Are respectively as follows:
wherein, wd i,j,k =w i,j,k -w i-1,j,k ,w i,j,k Is the value of the jth meteorological element at the kth time point of the historical day i,is the average of the jth meteorological factor of the historical day i.
Preferably, the similarity coefficient C of the history day i and the prediction day f is determined by the following formula fi :
C fi =ω 1 ·R fi ·ω 2
Wherein, ω is 1 Is a meteorological factor weight vector, omega 2 And (4) indicating a weight vector for the meteorological factors and the distance.
Preferably, performing ridge regression training on the training samples to obtain the prediction model comprises:
and (3) taking meteorological data in the training sample as independent variables, taking load data corresponding to the meteorological data in the training sample as dependent variables, and obtaining a prediction model by utilizing a ridge regression algorithm.
Further, after obtaining the prediction model, evaluating the prediction model using sample data:
taking meteorological data of each historical day in the training sample as input of the prediction model, and obtaining a load data prediction value of each historical day;
determining a mean square error corresponding to the prediction model by utilizing an evaluation function according to the load data prediction value of each historical day and load data corresponding to meteorological data of each historical day in sample data;
and if the value of the mean square error is smaller than a set value, the prediction model is qualified.
Further, determining a mean square error MSE corresponding to the prediction model by the following formula:
wherein, observed i Predicted load data corresponding to weather data of historical day i in sample data i The load data prediction value of the historical day i is shown, and n is the total days of the historical day.
A short-term load prediction apparatus based on ridge regression, the apparatus comprising:
the acquisition unit is used for acquiring meteorological data of historical days and predicted days and load data corresponding to the meteorological data of the historical days;
the determining unit is used for determining a training sample according to the similarity index value matrix of the meteorological data of the historical days and the forecast days and the load data corresponding to the meteorological data of the historical days;
and the prediction unit is used for inputting the meteorological data of the prediction day into a prediction model and determining the load data of the prediction day, wherein the prediction model is obtained by performing ridge regression training on the training samples.
Compared with the prior art, the invention has the following beneficial effects:
the method and the device of the invention obtain the meteorological data of the historical days and the forecast days and the load data corresponding to the meteorological data of the historical days; determining a training sample according to a similarity index value matrix of the meteorological data of the historical day and the forecast day and load data corresponding to the meteorological data of the historical day; the technical scheme of inputting the meteorological data of the forecast day into the forecast model to determine the load data of the forecast day can improve the precision and accuracy of short-term load forecast.
The method and the device also obtain a more accurate prediction model finally based on the training of irrelevant sample interference; the method specifically comprises the following steps: taking meteorological data in a training sample as independent variables, taking load data corresponding to the meteorological data in the training sample as dependent variables, and obtaining a prediction model by utilizing a ridge regression algorithm; and the mean square error of the prediction model is calculated by utilizing an evaluation function according to the load data prediction value of the historical day and the corresponding load data, and the load prediction is carried out only when the accuracy of the model is qualified, so that the high precision of the prediction model and the accuracy of a load prediction result are ensured.
Drawings
FIG. 1 is a flow chart of a short term load prediction method based on ridge regression according to the present invention;
FIG. 2 is a flow chart of obtaining training samples in an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a short-term load prediction apparatus based on ridge regression according to the present invention.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings.
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Most of the existing load prediction algorithms in the field adopt an empirical method, and the problem of low prediction accuracy exists by taking weather forecast data as samples, and in addition, the load prediction algorithms are not enough to meet the requirement of mass data mining.
The invention provides a short-term load prediction method based on ridge regression, which takes a sample obtained by screening historical sample data by a meteorological similarity coefficient as a training sample of a training prediction model, and utilizes the sample data to evaluate the prediction model obtained by training; the training model obtained based on the training of the sample without irrelevant sample interference has higher accuracy; the prediction model is used for predicting the load data of the transformer area in the future time period after being evaluated by the evaluation function, so that the accuracy of the prediction result is further ensured.
As shown in fig. 1, the method includes:
101. acquiring meteorological data of historical days and predicted days and load data corresponding to the meteorological data of the historical days;
102. determining a training sample according to a similarity index value matrix of the meteorological data of the historical day and the forecast day and load data corresponding to the meteorological data of the historical day;
103. and inputting the meteorological data of the prediction day into a prediction model, and determining the load data of the prediction day, wherein the prediction model is obtained by performing ridge regression training on training samples.
The weather data and the corresponding load data of the historical days are obtained by preprocessing historical original data; the historical original data comprises characteristic data such as load, meteorological information, holidays, dates and the like, and the days of the historical days relative to the predicted days can be accurately judged through the characteristic data such as the holidays, the dates and the like;
the pretreatment comprises the steps of carrying out the operation of cleaning and cleaning on historical original data;
the meteorological data in step 101 includes: temperature, humidity, air pressure, wind speed data;
for example: training sample characterization data are shown in table 1:
TABLE 1 training sample data Table
Specifically, according to the dates with the same weather type and weather change condition, the load and weather factors of the dates with the same weather type and weather change condition and the similarity of the load change condition of the previous day, the historical days with higher weather similarity to the predicted days are searched as training samples, and the prediction model is trained on the basis of the training samples, so that the interference of irrelevant samples on the establishment of the prediction model can be eliminated to the greatest extent, and the accuracy and pertinence of the final prediction model are improved.
In the above process, for the training samples, all dates before the prediction date are generally selected without distinction as the training samples, so that the final model trained in this way lacks the pertinence to the specific prediction date. Considering that the weather condition of the prediction date of each specific value of the unknown load can be generally obtained from meteorological prediction data issued by a meteorological department, aiming at a specific prediction day, training samples used for prediction can be firstly screened according to the meteorological condition, the screened training samples are adaptively selected according to the prediction day, targeted model training is realized, and finally a more accurate prediction model without irrelevant sample interference is obtained.
The method specifically comprises the following steps: for training samples obtained by preprocessing original data, the samples are screened through weather similarity indexes, the indexes comprehensively consider Euclidean distance of weather factors of a forecast day and a historical day, weather change conditions of a previous day, correlation coefficients and the like, and include various weather factors such as temperature, humidity, air pressure, air speed and the like.
Specifically, as shown in fig. 2, step 102 specifically includes:
determining a similarity coefficient between the historical day and the predicted day according to a similarity index value matrix of the meteorological data of the historical day and the predicted day;
and selecting the meteorological data of the historical day corresponding to the previous n large similarity coefficients in the similarity coefficients and the load data corresponding to the meteorological data as training samples.
Wherein, a similarity index value matrix R of the meteorological data of the historical day i and the predicted day f is determined by the following formula fi :
R fi =[r i1j ,r i2j ,r i3j ,r i4j ] T
Wherein r is i1j For the historical day i to the predicted day fEuclidean distance of j meteorological factors, r i2j The correlation coefficient r of the historical day i to the jth meteorological factor of the predicted day i3j The Euclidean distance r is obtained by first-order difference between the jth meteorological factor of the historical day i and the predicted day and the meteorological factor value of the previous day i4j The correlation coefficient is obtained by first-order difference between the historical day i and the jth meteorological factor of the forecast day and the meteorological factor value of the previous day, i belongs to [1, n ∈],j∈[1,m]N is the total days of the historical days, and m is the total number of meteorological factors in meteorological data;
r i1j 、r i2j 、r i3j and r i4j Are respectively as follows:
wherein, wd i,j,k =w i,j,k -w i-1,j,k ,w i,j,k Is the value of the jth meteorological factor at the kth time point of the historical day i,is the average of the jth meteorological factor of the historical day i.
Determining the similarity coefficient C between the historical day i and the predicted day f by the following formula fi :
C fi =ω 1 ·R fi ·ω 2
Wherein, ω is 1 Is a meteorological factor weight vector, omega 2 Distance being a meteorological factorAn index weight vector.
Performing ridge regression training on training samples to obtain the prediction model comprises the following steps:
taking meteorological data in a training sample as independent variables, taking load data corresponding to the meteorological data in the training sample as dependent variables, and obtaining a prediction model by utilizing a ridge regression algorithm;
specifically, the above training operation may be performed multiple times by using multiple sets of data, so as to adaptively adjust and update the prediction model.
Based on the steps, after obtaining the prediction model, evaluating the prediction model by using sample data:
the meteorological data of each historical day in the training sample is used as the input of the prediction model, and the load data prediction value of each historical day is obtained;
determining a mean square error corresponding to the prediction model by utilizing an evaluation function according to the load data prediction value of each historical day and the load data corresponding to the meteorological data of each historical day in the sample data; if the value of the mean square error is smaller than a set value, the prediction model is qualified;
specifically, the mean square error MSE corresponding to the prediction model is determined by the following formula:
wherein, observed i Predicted load data corresponding to weather data of historical day i in sample data i The load data prediction value of the historical day i is shown, and n is the total days of the historical day.
The mean square error is a loss function used as a least squares regression, representing the average of the squared differences of the predicted and actual values of all samples. The MSE evaluates the change degree of the data, and the smaller the value of the MSE is, the better accuracy of the prediction model describing the experimental data is shown; the input data of the evaluation function MSE can be set for multiple times to repeat evaluation operation, and an optimal prediction model is determined.
Fig. 3 is a short-term load prediction apparatus based on ridge regression according to an embodiment of the present invention, including:
the acquisition unit is used for acquiring meteorological data of historical days and predicted days and load data corresponding to the meteorological data of the historical days;
the determining unit is used for determining a training sample according to the similarity index value matrix of the meteorological data of the historical days and the forecast days and the load data corresponding to the meteorological data of the historical days;
and the prediction unit is used for inputting the meteorological data of the prediction day into a prediction model and determining the load data of the prediction day, wherein the prediction model is obtained by performing ridge regression training on the training samples.
The determination unit includes: the system comprises a data preprocessing module, a calculation module, a data screening module and an evaluation module;
the data preprocessing module is used for preprocessing historical raw data and comprises: aligning historical original data and cleaning historical original data;
a calculation module: the similarity index value matrix is used for determining the similarity coefficient between the historical day and the predicted day according to the weather data similarity index value matrix between the historical day and the predicted day;
the data screening module: the system is used for screening training samples according to the similarity coefficient of the historical day and the prediction day;
an evaluation module: and the evaluation function is used for determining the mean square error corresponding to the prediction model and determining whether the prediction model is qualified.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (7)
1. A method for short-term load prediction based on ridge regression, the method comprising:
acquiring meteorological data of historical days and predicted days and load data corresponding to the meteorological data of the historical days;
determining a training sample according to a similarity index value matrix of the meteorological data of the historical day and the forecast day and load data corresponding to the meteorological data of the historical day;
inputting meteorological data of a prediction day into a prediction model, and determining load data of the prediction day, wherein the prediction model is obtained by performing ridge regression training on training samples;
the determining of the training sample according to the similarity index value matrix of the meteorological data of the historical day and the forecast day and the load data corresponding to the meteorological data of the historical day comprises the following steps:
determining a similarity coefficient between the historical day and the predicted day according to the similarity index value matrix of the meteorological data of the historical day and the predicted day;
selecting meteorological data of a historical day corresponding to the previous n large similarity coefficients in the similarity coefficients and load data corresponding to the meteorological data as training samples;
determining a similarity index value matrix R of meteorological data of the historical day i and the predicted day f by the following formula fi :
R fi =[r i1j ,r i2j ,r i3j ,r i4j ] T
Wherein r is i1j Euclidean distance, r, of the jth meteorological factor for historical day i versus predicted day f i2j The correlation coefficient r of the historical day i to the jth meteorological factor of the predicted day i3j The Euclidean distance r is obtained by first-order difference between the jth meteorological factor of the historical day i and the predicted day and the meteorological factor value of the previous day i4j The correlation coefficient is obtained by first-order difference between the historical day i and the jth meteorological factor of the forecast day and the meteorological factor value of the previous day, i belongs to [1, n ∈],j∈[1,m]N is the total days of the historical days, and m is the total number of meteorological factors in meteorological data;
said r i1j 、r i2j 、r i3j And r i4j Are respectively as follows:
2. The method of claim 1, wherein the meteorological data comprises: temperature, humidity, air pressure, wind speed data.
3. The method of claim 1 wherein the similarity coefficient C between the historical day i and the predicted day f is determined by fi :
C fi =ω 1 ·R fi ·ω 2
Wherein, ω is 1 Is a meteorological factor weight vector, omega 2 And (4) indicating a weight vector for the meteorological factors and the distance.
4. The method of claim 1, wherein performing ridge regression training on training samples to obtain the prediction model comprises:
and (3) taking meteorological data in the training sample as independent variables, taking load data corresponding to the meteorological data in the training sample as dependent variables, and obtaining a prediction model by using a ridge regression algorithm.
5. The method of claim 4, wherein the predictive model is evaluated using sample data after it is obtained:
taking meteorological data of each historical day in the training sample as input of the prediction model, and obtaining a load data prediction value of each historical day;
determining a mean square error corresponding to the prediction model by utilizing an evaluation function according to the load data prediction value of each historical day and load data corresponding to meteorological data of each historical day in sample data;
and if the value of the mean square error is smaller than a set value, the prediction model is qualified.
6. The method of claim 5, wherein the Mean Square Error (MSE) corresponding to the prediction model is determined by:
wherein, observed i Predicted load data corresponding to weather data of historical day i in sample data i The load data prediction value of the historical day i is shown, and n is the total days of the historical day.
7. A ridge-regression-based short-term load prediction apparatus for use in the ridge-regression-based short-term load prediction method according to any one of claims 1 to 6, the apparatus comprising:
the acquisition unit is used for acquiring meteorological data of historical days and predicted days and load data corresponding to the meteorological data of the historical days;
the determining unit is used for determining a training sample according to the similarity index value matrix of the meteorological data of the historical days and the forecast days and the load data corresponding to the meteorological data of the historical days;
and the prediction unit is used for inputting the meteorological data of the prediction day into a prediction model and determining the load data of the prediction day, wherein the prediction model is obtained by performing ridge regression training on the training samples.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710690551.9A CN109389238B (en) | 2017-08-14 | 2017-08-14 | Ridge regression-based short-term load prediction method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710690551.9A CN109389238B (en) | 2017-08-14 | 2017-08-14 | Ridge regression-based short-term load prediction method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109389238A CN109389238A (en) | 2019-02-26 |
CN109389238B true CN109389238B (en) | 2022-08-09 |
Family
ID=65416927
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710690551.9A Active CN109389238B (en) | 2017-08-14 | 2017-08-14 | Ridge regression-based short-term load prediction method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109389238B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110222727A (en) * | 2019-05-15 | 2019-09-10 | 广东电网有限责任公司电力调度控制中心 | A kind of short-term load forecasting method and device based on deep neural network |
CN110648248B (en) * | 2019-09-05 | 2023-04-07 | 广东电网有限责任公司 | Control method, device and equipment for power station |
CN111027193A (en) * | 2019-12-02 | 2020-04-17 | 浙江工业大学 | Short-term water level prediction method based on regression model |
CN112700043B (en) * | 2020-12-31 | 2024-03-08 | 新奥数能科技有限公司 | Load prediction method, load prediction device, computer readable storage medium and electronic equipment |
CN113159426A (en) * | 2021-04-25 | 2021-07-23 | 中科三清科技有限公司 | Weather type similarity judgment method and device, electronic equipment and readable storage medium |
CN114243695B (en) * | 2021-12-17 | 2023-09-15 | 中国矿业大学(北京) | Power load prediction method based on bidirectional long-short-term memory neural network |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106408223A (en) * | 2016-11-30 | 2017-02-15 | 华北电力大学(保定) | Short-term load prediction based on meteorological similar day and error correction |
CN106779129A (en) * | 2015-11-19 | 2017-05-31 | 华北电力大学(保定) | A kind of Short-Term Load Forecasting Method for considering meteorologic factor |
CN107016469A (en) * | 2017-04-13 | 2017-08-04 | 重庆大学 | Methods of electric load forecasting |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7403928B2 (en) * | 2005-02-03 | 2008-07-22 | Siemens Power Transmission & Distribution, Inc. | Identify data sources for neural network |
-
2017
- 2017-08-14 CN CN201710690551.9A patent/CN109389238B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106779129A (en) * | 2015-11-19 | 2017-05-31 | 华北电力大学(保定) | A kind of Short-Term Load Forecasting Method for considering meteorologic factor |
CN106408223A (en) * | 2016-11-30 | 2017-02-15 | 华北电力大学(保定) | Short-term load prediction based on meteorological similar day and error correction |
CN107016469A (en) * | 2017-04-13 | 2017-08-04 | 重庆大学 | Methods of electric load forecasting |
Non-Patent Citations (1)
Title |
---|
基于相似日选取的小波极限学习机短期负荷预测模型研究;谈力;《《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20160215;第26-64页 * |
Also Published As
Publication number | Publication date |
---|---|
CN109389238A (en) | 2019-02-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109389238B (en) | Ridge regression-based short-term load prediction method and device | |
CN111932402B (en) | Short-term power load bidirectional combination prediction method based on similar day and LSTM | |
CN107505837A (en) | A kind of semi-supervised neural network model and the soft-measuring modeling method based on the model | |
CN106779129A (en) | A kind of Short-Term Load Forecasting Method for considering meteorologic factor | |
CN102183621A (en) | Aquaculture dissolved oxygen concentration online forecasting method and system | |
CN110332605B (en) | BP neural network-based heat supply network energy consumption index analysis and early warning method | |
CN105740988A (en) | Prediction method of coal calorific value on the basis of grey correlation analysis and multiple linear regression model | |
CN106600037B (en) | Multi-parameter auxiliary load prediction method based on principal component analysis | |
CN108334943A (en) | The semi-supervised soft-measuring modeling method of industrial process based on Active Learning neural network model | |
CN110544180A (en) | Building energy-saving system based on energy consumption prediction and analysis diagnosis | |
CN111028100A (en) | Refined short-term load prediction method, device and medium considering meteorological factors | |
CN113837488B (en) | Method, system and equipment for predicting energy consumption data | |
CN110826855A (en) | Method and system for testing network access performance of intelligent power distribution room state monitoring sensor | |
CN109583645A (en) | A kind of public building short-term load forecasting method | |
CN107633316A (en) | The power network short-term load forecasting method of adaptive local nonlinear regression deviation compensation | |
CN110533247B (en) | Monthly power consumption prediction method adopting temperature data anomaly point compensation | |
CN116090821A (en) | Power distribution network line security risk assessment method considering extreme weather | |
CN110737975B (en) | Wind power plant wind speed and power prediction and abnormality correction method based on empirical mode decomposition and autoregressive model | |
CN112348264A (en) | Carbon steel corrosion rate prediction method based on random forest algorithm | |
CN110298765B (en) | Power distribution network power consumption abnormality detection method based on objective correlation factors | |
CN115511100A (en) | Air conditioner load regression prediction method based on environmental temperature related data learning | |
CN115358437A (en) | Power supply load prediction method based on convolutional neural network | |
CN108830405B (en) | Real-time power load prediction system and method based on multi-index dynamic matching | |
CN111091223A (en) | Distribution transformer short-term load prediction method based on Internet of things intelligent sensing technology | |
CN111027841A (en) | Low-voltage transformer area line loss calculation method based on gradient lifting decision tree |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |