CN114022021A - Online unit load prediction method based on ensemble learning - Google Patents
Online unit load prediction method based on ensemble learning Download PDFInfo
- Publication number
- CN114022021A CN114022021A CN202111357431.XA CN202111357431A CN114022021A CN 114022021 A CN114022021 A CN 114022021A CN 202111357431 A CN202111357431 A CN 202111357431A CN 114022021 A CN114022021 A CN 114022021A
- Authority
- CN
- China
- Prior art keywords
- unit load
- model
- load prediction
- prediction method
- features
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 19
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 230000002776 aggregation Effects 0.000 claims abstract description 6
- 238000004220 aggregation Methods 0.000 claims abstract description 6
- 238000012417 linear regression Methods 0.000 claims abstract description 6
- 238000012545 processing Methods 0.000 claims abstract description 4
- 238000000605 extraction Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims 1
- 238000003062 neural network model Methods 0.000 abstract 1
- 230000008859 change Effects 0.000 description 7
- 239000010410 layer Substances 0.000 description 6
- 230000006872 improvement Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 1
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 description 1
- 206010033307 Overweight Diseases 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005065 mining 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
- 210000002569 neuron Anatomy 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 239000002356 single layer Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000010977 unit operation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06313—Resource planning in a project environment
-
- 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)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Game Theory and Decision Science (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Biodiversity & Conservation Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an on-line unit load prediction method based on ensemble learning, which comprises the following steps: reading historical data of unit load in past month from a database, preprocessing the data, detecting a coarse value according to a Lauder criterion, and processing the detected coarse value in a backward filling mode. Further extracting features of the preprocessed data, wherein the features comprise time features, correlation features and aggregation features, and constructing a feature data set; training a limit gradient lifting model and a lightweight high-efficiency gradient lifting model through the feature data set, and training a long-short term memory neural network model through a prediction result of the LGBM model and original feature data; and finally, taking the output results of the three models as three inputs to train a linear regression model to output the final load prediction result. And in the model training stage, network parameters are adjusted on the principle that the average prediction error is minimum, so that an optimal unit load prediction model is established.
Description
Technical Field
The invention belongs to the field of thermal power station load prediction, and relates to an on-line unit load prediction method based on ensemble learning.
Background
Under the influence of factors such as environment, seasons and the like, the power demand of a power grid fluctuates greatly along with time, and the balance relation of a supply end and a demand end needs to be maintained according to the actual characteristics of power production and consumption. The generated power generation amount change of the power supply side is mainly adjusted by the thermal power generating unit through the change of the unit load, and a load plan curve is issued to a power plant by a power grid according to the actual condition of the unit and a corresponding power dispatching principle, but the current unit grid-connected structure is complex, and the plan curve has larger deviation with the actual unit load instruction. Therefore, accurate unit actual load prediction is helpful for a power plant to select an adjusting mode, timely responds to load scheduling of a power grid, and can enable the power grid to keep running efficiently, stably, safely and economically.
Because the load demand is influenced by various condition factors, the load demand has the characteristics of uncertainty, time periodicity, conditionality and the like, the following load prediction methods are mainly adopted at present: (1) performing regression prediction; (2) predicting a time sequence; (3) predicting gray; (4) an artificial neural network; (5) a support vector machine; (6) a wavelet change algorithm; (7) and (5) fuzzy regression model. However, the non-linearity of the coefficients in the regression method reduces the accuracy of the model; the time prediction model is greatly influenced by factors such as weather and weather; the prediction accuracy of the gray model is inversely proportional to the gray scale, and when the data discrete degree is increased, the prediction accuracy is reduced; the design requirement of the artificial neural network is high, the number of hidden layers is difficult to judge, and the convergence speed is low; the support vector machine is difficult to process large-scale training samples and cannot reflect the change rule of the unit load for a long time; the wavelet transformation algorithm process is complex, the requirement on the accuracy of historical load data is high, and the interference effect of complex factors is difficult to overcome; the fuzzy system lacks self-learning ability, the fuzzy rule of the fuzzy system mainly depends on an expert system, and the application range has larger limitation.
Therefore, in order to respond the load scheduling requirement of the generator set in time, a load prediction method which is suitable for the load characteristic of the generator set and has higher prediction precision is needed urgently at present, and a stable and accurate time sequence change trend is obtained from complex and disordered load operation data through an effective data processing and mining tool.
Disclosure of Invention
In order to quickly respond to a load scheduling instruction, the invention provides an online load prediction method based on ensemble learning, which judges the change trend of a load by predicting and judging the result and determines a load regulation mode.
In order to solve the technical problems, the invention adopts the following technical scheme:
an online unit load prediction method based on ensemble learning comprises the following steps:
(1) acquiring historical load operation data of the previous month from a unit SIS database to obtain an original data sample D;
(2) carrying out data preprocessing on an original data sample D, removing a gross error according to a Lauda criterion, and recording a processed sample as CD;
(3) performing feature extraction on the sample CD, including time feature correlation features and aggregation features, and establishing a feature data sample TD;
(4) inputting a TD input limit gradient lifting model and a lightweight high-efficiency gradient lifting model, and respectively recording corresponding output results as XD and LD 1;
(5) splicing TD and LD1 to obtain a full connection layer Dense1, inputting the full connection layer Dense1 into a long-short term memory neural network, and recording the output result as LD 2;
(6) and finally, splicing the XD, the LD1 and the LD2 to be used as a full connection layer Dense2, inputting the full connection layer Dense2 into a linear regression model, and obtaining a final output result, namely a final unit load prediction result.
The invention is further improved in that the sampling interval in the step (1) is 1 min.
The invention is further improved in that the historical data in the step (1) is time sequence historical data of unit load.
In a further improvement of the present invention, the data preprocessing process in step (2) is to filter outliers in the original sample D.
The invention is further improved in that in the step (2), the standard deviation is obtained by calculating and processing the sample D according to the Lauda criterion, the exceeding deviation interval is a coarse error, and the load value at the previous moment is used for replacing the error value.
The invention further improves the characteristic extraction in the step (3) to comprise discrete time characteristics, autocorrelation characteristics and partial autocorrelation characteristics between a current value and a past value thereof and aggregation characteristics of average values of different time windows, wherein the time span is 3, namely t-1, t-2 and t-3 are included.
In a further improvement of the present invention, the splicing in step (5) is matrix splicing of TD and LD, and does not add corresponding elements.
The invention is further improved in that in the step (6), the linear regression model linearly combines the results of XD, LD1 and LD2 through an addition model, and the weight value is continuously changed during training to reduce the prediction residual error so as to obtain a satisfactory training result.
Compared with the prior art, the invention has the following advantages:
(1) the invention is an algorithm-based online prediction method without additional hardware equipment.
(2) The unit load prediction method based on the running data and the fusion of multiple algorithms for integrated learning is high in prediction accuracy and stability, small in maximum prediction error in a long time period and stable.
(3) According to the method, a block structure is established by sequencing the models during model training through an extreme gradient boost (XGboost) method, repeated data in iterative computation are reduced, computation complexity is reduced, and computation time is shortened.
(4) According to the method, the weights of different historical data are calculated through a lightweight high-efficiency gradient boosting (LGBM) algorithm, the data with smaller weight in the modeling data are reduced, and the model accuracy is improved.
(5) According to the method, the proportion of high-weight data is enhanced through matrix splicing of multi-model output results, low-weight data are stored, the generalization capability of the model is improved on the basis of ensuring the accuracy of the model, and the root mean square error between an actual value and a predicted value of the model is reduced.
Drawings
FIG. 1 is a flowchart of an online unit load prediction method based on ensemble learning according to the present invention.
FIG. 2 is a time sequence variation curve of the unit load under the same working condition in the embodiment.
Fig. 3 is a comparison graph of the prediction result of the predicted unit load time sequence and the actual unit load in the embodiment.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art. It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict. The present invention will be described in detail below with reference to the embodiments with reference to the attached drawings.
The framework of the invention mainly comprises the core steps of historical data sampling, data preprocessing, feature extraction, XGboost model training, LGBM model training, LSTM model training, ensemble learning linear regression model training, unit load prediction and the like, and the detailed flow is shown in figure 1. Taking a certain coal-fired unit as an example, the method comprises the following specific operation steps:
load historical data values of 1/00/2020/5/1/00 to 31/00/2020/5/31 are collected from a database of a plant-level monitoring information system (SIS), the number taking interval is 1min, and the obtained load value sequence is shown in FIG. 2. And preprocessing the data, calculating a standard deviation interval according to a Lauda criterion, and cleaning a coarse error value. And further performing feature extraction on the data, dispersing time features, extracting autocorrelation features and partial autocorrelation features between a current value and a past value thereof and aggregation features of different time window average values, wherein the time span is 3 and comprises t-1, t-2 and t-3, dividing feature data into a data set according to a ratio of 4:1, taking the former part as training data and the latter part as test data. And further training two models of XGboost and LGBM respectively by using training data, wherein the learner trains by adopting four-fold cross validation, and parameter optimization is carried out by using a GridSearch method in a Sklearn class.
The training data was used to train the constructed predictive model, the parameters of which are shown in table 1.
Table 1: XGboost model parameters
| n_estimators | 16000 |
| min_child_weight | 1 |
| learning_rate | 0.01 |
| max_depth | 6 |
| subsample | 0.6 |
| colsample_bytree | 0.4 |
| colsample_bylevel | 1 |
Table 2: LGBM model parameters
| n_estimators | 9000 |
| |
20 |
| |
20 |
| learning_rate | 0.005 |
| feature_fraction | 0.8 |
| subsample | 0.4 |
The LGBM and the XGboost are both provided with early termination training: training is stopped when the model loss does not improve for 300 consecutive iterations. For the LSTM model, a single layer LSTM is used, with an implied layer neuron number of 200 and an iteration number of 200.
FIG. 3 shows the comparison between the unit load prediction result and the actual measuring point data in a period of time (9: 00 days 5/month in 2020 to 00:00 days 22/month in 2020). The Root Mean Square Error (RMSE) of the model predictions was 3.2796 and the absolute value of the relative percent error (MAPE) was 2.147% for the entire test set. As can be seen, the model prediction error is very small, and the requirements in practical engineering are met. Meanwhile, the actually measured unit load time sequence data has high volatility, the prediction result of the model can well identify the fluctuation rule of the load, and the model is well fit with the actually measured result, so that the unit load prediction method provided by the invention can judge the change trend of the load in advance, and the problem that the load is difficult to respond to the instruction quickly in the unit operation process is solved.
The above description is only of the preferred embodiments of the present invention, and it should be noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the invention and these are intended to be within the scope of the invention.
Claims (8)
1. An online unit load prediction method based on ensemble learning is characterized by comprising the following steps:
(1) acquiring historical load operation data of the previous month from a unit SIS database to obtain an original data sample D;
(2) carrying out data preprocessing on an original data sample D, removing a gross error according to a Lauda criterion, and recording a processed sample as CD;
(3) performing feature extraction on the sample CD, including time feature correlation features and aggregation features, and establishing a feature data sample TD;
(4) inputting a TD input limit gradient lifting model and a lightweight high-efficiency gradient lifting model, and respectively recording corresponding output results as XD and LD 1;
(5) splicing TD and LD1 to obtain a full connection layer Dense1, inputting the full connection layer Dense1 into a long-short term memory neural network, and recording the output result as LD 2;
(6) and finally, splicing the XD, the LD1 and the LD2 to be used as a full connection layer Dense2, inputting the full connection layer Dense2 into a linear regression model, and obtaining a final output result, namely a final unit load prediction result.
2. The online unit load prediction method based on ensemble learning as claimed in claim 1, wherein the sampling interval in step (1) is 1 min.
3. The online unit load prediction method based on ensemble learning as claimed in claim 1, wherein the historical data in step (1) is time-series historical data of the unit load.
4. The ensemble learning-based online unit load prediction method according to claim 1, wherein the data preprocessing process in the step (2) is to filter outliers in the original sample D.
5. The on-line unit load prediction method based on ensemble learning as claimed in claim 4, wherein in step (2), according to the Lauda criterion, the standard deviation is obtained by calculation processing on the sample D, the exceeding deviation interval is a gross error, and the load value at the previous moment is used to replace the error value.
6. The ensemble learning-based online unit load prediction method according to claim 1, wherein the feature extraction in step (3) includes discrete time features, autocorrelation features and partial autocorrelation features between a current value and a past value thereof, and aggregation features of different time window averages, wherein each time span is 3, that is, t-1, t-2, and t-3 are included.
7. The ensemble learning-based online unit load prediction method according to claim 1, wherein the splicing in the step (5) is a matrix splicing of TD and LD, and the non-corresponding elements are added.
8. The on-line unit load prediction method based on ensemble learning as claimed in claim 1, wherein in the step (6), the linear regression model linearly combines the XD, LD1 and LD2 results through an addition model, and the weight is continuously changed during training to reduce the prediction residual so as to obtain a satisfactory training result.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111357431.XA CN114022021A (en) | 2021-11-16 | 2021-11-16 | Online unit load prediction method based on ensemble learning |
| PCT/CN2022/131741 WO2023088212A1 (en) | 2021-11-16 | 2022-11-14 | Online unit load prediction method based on ensemble learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111357431.XA CN114022021A (en) | 2021-11-16 | 2021-11-16 | Online unit load prediction method based on ensemble learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114022021A true CN114022021A (en) | 2022-02-08 |
Family
ID=80064606
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111357431.XA Pending CN114022021A (en) | 2021-11-16 | 2021-11-16 | Online unit load prediction method based on ensemble learning |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114022021A (en) |
| WO (1) | WO2023088212A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114444821A (en) * | 2022-04-12 | 2022-05-06 | 国网湖北省电力有限公司电力科学研究院 | Integrated learning load prediction method, system and medium for power internet of things |
| WO2023088212A1 (en) * | 2021-11-16 | 2023-05-25 | 西安热工研究院有限公司 | Online unit load prediction method based on ensemble learning |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116544931B (en) * | 2023-06-27 | 2023-12-01 | 北京理工大学 | Power load distribution prediction method based on integrated fragment transformation and time convolution network |
| CN116632842B (en) * | 2023-07-26 | 2023-11-10 | 国网山东省电力公司信息通信公司 | Clustering characteristic-based method and system for predicting distribution type photovoltaic load probability of platform |
| CN117239731B (en) * | 2023-09-21 | 2024-02-27 | 山东工商学院 | Holiday short-term power load prediction method based on hybrid model |
| CN117078047B (en) * | 2023-10-16 | 2024-02-23 | 华能济南黄台发电有限公司 | LSTM-based heat load prediction and distribution optimization method and system |
| CN117909928B (en) * | 2024-03-19 | 2024-05-28 | 国网四川省电力公司成都供电公司 | Air conditioner load prediction method and system based on big data analysis |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9639643B2 (en) * | 2014-06-06 | 2017-05-02 | General Electric Company | Method and system for generating electric load models |
| CN110084424A (en) * | 2019-04-25 | 2019-08-02 | 国网浙江省电力有限公司 | A kind of Methods of electric load forecasting based on LSTM and LGBM |
| CN113205207A (en) * | 2021-04-19 | 2021-08-03 | 深圳供电局有限公司 | XGboost algorithm-based short-term power consumption load fluctuation prediction method and system |
| CN114022021A (en) * | 2021-11-16 | 2022-02-08 | 西安热工研究院有限公司 | Online unit load prediction method based on ensemble learning |
-
2021
- 2021-11-16 CN CN202111357431.XA patent/CN114022021A/en active Pending
-
2022
- 2022-11-14 WO PCT/CN2022/131741 patent/WO2023088212A1/en unknown
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023088212A1 (en) * | 2021-11-16 | 2023-05-25 | 西安热工研究院有限公司 | Online unit load prediction method based on ensemble learning |
| CN114444821A (en) * | 2022-04-12 | 2022-05-06 | 国网湖北省电力有限公司电力科学研究院 | Integrated learning load prediction method, system and medium for power internet of things |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023088212A1 (en) | 2023-05-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114022021A (en) | Online unit load prediction method based on ensemble learning | |
| CN108388962B (en) | Wind power prediction system and method | |
| CN112149879B (en) | New energy medium-and-long-term electric quantity prediction method considering macroscopic volatility classification | |
| CN108053082B (en) | Power grid medium and long term load prediction method based on temperature interval decomposition | |
| CN108764568B (en) | Data prediction model tuning method and device based on LSTM network | |
| CN111241755A (en) | Power load prediction method | |
| CN109615124B (en) | SCADA master station load prediction method based on deep learning | |
| CN112926770A (en) | Unified data metering and collecting system based on digital twins | |
| CN115587672A (en) | Distribution transformer load prediction and heavy overload early warning method and system | |
| CN110197296B (en) | Unit load prediction method based on time series similarity | |
| CN116167527B (en) | Pure data-driven power system static safety operation risk online assessment method | |
| CN113988481B (en) | Wind power prediction method based on dynamic matrix prediction control | |
| CN113919545A (en) | Photovoltaic power generation power prediction method and system with integration of multiple data models | |
| CN112085285A (en) | Bus load prediction method and device, computer equipment and storage medium | |
| CN111340305A (en) | Building operation energy consumption prediction method | |
| CN111754034A (en) | Time sequence prediction method based on chaos optimization neural network model | |
| CN115860177A (en) | Photovoltaic power generation power prediction method based on combined machine learning model and application thereof | |
| CN116167531A (en) | Photovoltaic power generation prediction method based on digital twin | |
| CN115222106A (en) | User day-ahead load prediction method of self-adaptive model | |
| CN115115127A (en) | Low-voltage transformer area intelligent regulation and control method and system based on deep learning | |
| CN116014722A (en) | Sub-solar photovoltaic power generation prediction method and system based on seasonal decomposition and convolution network | |
| CN112232570A (en) | Forward active total electric quantity prediction method and device and readable storage medium | |
| CN116562454A (en) | Manufacturing cost prediction method applied to BIM long-short-time attention mechanism network | |
| CN116191412A (en) | Power load prediction method | |
| CN116402194A (en) | Multi-time scale load prediction method based on hybrid neural network |
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 |