CN111178417A - Energy accurate load prediction method for individual and group of users - Google Patents
Energy accurate load prediction method for individual and group of users Download PDFInfo
- Publication number
- CN111178417A CN111178417A CN201911337331.3A CN201911337331A CN111178417A CN 111178417 A CN111178417 A CN 111178417A CN 201911337331 A CN201911337331 A CN 201911337331A CN 111178417 A CN111178417 A CN 111178417A
- Authority
- CN
- China
- Prior art keywords
- load
- data
- prediction
- adopting
- weights
- 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 17
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000012360 testing method Methods 0.000 claims abstract description 4
- 238000012549 training Methods 0.000 claims description 21
- 230000000694 effects Effects 0.000 claims description 9
- 230000005611 electricity Effects 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000009499 grossing Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 description 4
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 238000004378 air conditioning Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004941 influx Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Marketing (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Probability & Statistics with Applications (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention provides an energy accurate load prediction method for individual and group users, which comprises the following steps: s1, preprocessing the data of the power load; s2, classifying the load scenes by adopting an improved k-means algorithm according to the load influencing factors; s3, inputting the influence load factors and the historical load data into an ELM model by adopting an extreme learning machine ELM to obtain a preliminary load predicted value; s4, establishing a strong learner with high prediction precision by adopting an Adaboost algorithm, and correcting the preliminary load prediction value; s5, continuously learning through a strong learner, improving the accuracy of load prediction and finally obtaining an accurate load test value; the method has the advantage of accurate load prediction.
Description
Technical Field
The invention belongs to the technical field of power distribution networks, and particularly relates to an energy accurate load prediction method for individual users and groups.
Background
With the rapid growth of population and the rapid development of economy, the electricity consumption of residential buildings is remarkably increased. In order to balance the demand for electricity and reduce carbon emissions, the development of smart buildings and smart grids is receiving more and more attention. Meanwhile, the intermittency and the volatility of the renewable energy bring certain influence on the power grid. With the increase of intelligent electricity utilization terminals in residential houses, the electricity utilization of residents has stronger fluctuation and randomness, so that the balance of supply and demand of electric power can be influenced. Therefore, reliable and accurate load prediction has important significance, and the method is favorable for realizing dynamic planning and efficient management of intelligent buildings and intelligent power grids and improving the utilization efficiency of renewable energy sources. In many tourist cities, the load of the power distribution network has the typical characteristics of obvious valley-peak fluctuation, concentrated peak periods and intermittent overload, and the typical load represented by air conditioners, lighting facilities and the like has the characteristic of extremely rapid increase along with the mass influx of tourists in the peak periods of tourism; in extreme weather conditions, certain loads such as air conditioning may also exhibit a tendency to explosive surge. Typical analysis is carried out on the electricity utilization habits and consumption characteristics of individual users and groups of users, the typical characteristics of loads such as inns, hotels, restaurants, businesses and the like are considered in a focused mode, response modes of the users are researched, and response models are constructed.
Disclosure of Invention
The invention aims to provide an energy accurate load prediction method for individual and group users, and aims to solve the problems that the load of a power grid is changed greatly in different periods of a tourist city, the load prediction is inaccurate in the existing method, and the stable operation of the power grid is difficult to ensure.
The invention provides the following technical scheme:
an energy accurate load prediction method for user individuals and groups comprises the following steps: s1, preprocessing the data of the power load; s2, classifying the load scenes by adopting an improved k-means algorithm according to the load influencing factors; s3, inputting the influence load factors and the historical load data into an ELM model by adopting an extreme learning machine ELM to obtain a preliminary load predicted value; s4, establishing a strong learner with high prediction precision by adopting an Adaboost algorithm, and correcting the preliminary load prediction value; and S5, continuously learning by a strong learner, improving the accuracy of load prediction and finally obtaining an accurate load test value.
Further, in S1, the data preprocessing includes filling missing data, correcting noise data, smoothing data, and normalizing data.
Further, in S2, the load influencing factors include electricity prices, weather data and time variables; the load scenes include customer stacks, hotels, restaurants, and business load scenes.
Further, in S2, the improved k-means algorithm includes giving a cluster number k, randomly selecting k centroid points, and assigning the most similar class to each load scene according to the similarity between each load scene and each centroid point.
Further, in S3, the extreme learning machine ELM is a fast learning algorithm, and for the single hidden layer neural network, the ELM model may randomly initialize the input weights and the offsets and obtain the corresponding output weights.
Further, in S4, the strong learner assigns a higher weight to the sample with poor training effect in the training set, and appropriately increases the weight of the learner with strong learning ability and good training effect, and accordingly reduces the weight of the sample with good training effect and the weight of the learner with poor learning ability.
The invention has the beneficial effects that:
according to the energy accurate load prediction method for individual users and groups, accurate load prediction values are obtained through data preprocessing and integrated learning, so that dynamic planning and efficient management of intelligent buildings and intelligent power grids are achieved, and meanwhile, the utilization efficiency of renewable energy sources is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a schematic flow diagram of the present invention;
fig. 2 is a schematic structural diagram of the present invention.
Detailed Description
As shown in fig. 1-2, a method for predicting energy accurate load of individual and group users includes the following steps:
s1, preprocessing the data of the power load;
the data preprocessing comprises filling missing data, correcting noise data, smoothing data and normalizing data;
s2, classifying the load scenes by adopting an improved k-means algorithm according to the load influencing factors;
load influencing factors include electricity prices, weather data and time variables; the load scene comprises a hotel, a restaurant and a business load scene; the improved k-means algorithm comprises the steps of giving a clustering number k, randomly selecting k centroid points, and assigning the centroid points to the most similar classes according to the similarity of each load scene and each centroid point;
s3, inputting the influence load factors and the historical load data into an ELM model by adopting an extreme learning machine ELM to obtain a preliminary load predicted value;
the ELM is a fast learning algorithm, and for a single hidden layer neural network, an ELM model can randomly initialize input weight and bias and obtain corresponding output weight;
s4, establishing a strong learner with high prediction precision by adopting an Adaboost algorithm, and correcting the preliminary load prediction value; the strong learner gives a higher weight to the samples with poorer training effect in the training set, and appropriately increases the weight of the learner with strong learning ability and good training effect, and correspondingly reduces the weight of the samples with better training effect and the weight of the learner with weaker learning ability;
the Adaboost algorithm is the most popular of the current ensemble learning algorithms, and allows new sub-learners to be added continuously until the prediction accuracy requirement is met. The prediction accuracy can be sufficiently high as long as the sub-learners are sufficiently large. In the Adaboost algorithm, each training sample in the original training sample set is given a weight to reflect the importance of the sample, which represents the probability that the sample can be selected into a certain training subset. If a sample has been accurately predicted, its weight is reduced and the probability of being selected into the training subset is reduced. In this way, the sub-learner may be emphatically trained on samples that have not been correctly predicted. The training error of the final prediction function h of the Adaboost algorithm satisfies:
H＝Π[2∈i(1-∈i)]＝Π1-4(12-∈i)≤exp(-2Σi(12-∈i))
wherein: h is a prediction function; epsilon i is the prediction error of the individual learner hi obtained by training.
This equation shows that as long as the training error ε i of the individual learner is slightly better than the random guess, i.e., ε i < 0.5, the training error of the final predictor function H decreases exponentially with i.
Although the Adaboost algorithm is proved by theory that the training error can be small enough as long as the number i of the sub-learners is enough. However, in practice, it is necessary to obtain a prediction function with a simple model and satisfactory accuracy in a short time based on efficiency considerations. And the proper sub-learner is selected, so that the complexity of Adaboost algorithm operation can be reduced, and the prediction speed is greatly improved.
And S5, continuously learning by a strong learner, improving the accuracy of load prediction and finally obtaining an accurate load test value.
Although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. An energy accurate load prediction method for individual and group users is characterized by comprising the following steps:
s1, preprocessing the data of the power load;
s2, classifying the load scenes by adopting an improved k-means algorithm according to the load influencing factors;
s3, inputting the influence load factors and the historical load data into an ELM model by adopting an extreme learning machine ELM to obtain a preliminary load predicted value;
s4, establishing a strong learner with high prediction precision by adopting an Adaboost algorithm, and correcting the preliminary load prediction value;
and S5, continuously learning by a strong learner, improving the accuracy of load prediction and finally obtaining an accurate load test value.
2. The method of claim 1, wherein in step S1, the data preprocessing includes filling missing data, correcting noise data, smoothing data, and normalizing data.
3. The method according to claim 1, wherein in S2, the factors affecting load include electricity price, weather data and time variation; the load scenes include customer stacks, hotels, restaurants, and business load scenes.
4. The method of claim 1, wherein in step S2, the improved k-means algorithm comprises giving a cluster number k, randomly selecting k centroid points, and assigning the most similar class according to the similarity between each load scene and each centroid point.
5. The method of claim 1, wherein in step S3, the extreme learning machine ELM is a fast learning algorithm, and for the single hidden layer neural network, the ELM model randomly initializes input weights and biases and obtains corresponding output weights.
6. The method according to claim 1, wherein in S4, the strong learner assigns higher weights to the samples with poor training effect in the training set, and appropriately increases the weights of the learners with strong learning ability and good training effect, and accordingly reduces the weights of the samples with good training effect and the learners with weak learning ability.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911337331.3A CN111178417A (en) | 2019-12-23 | 2019-12-23 | Energy accurate load prediction method for individual and group of users |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911337331.3A CN111178417A (en) | 2019-12-23 | 2019-12-23 | Energy accurate load prediction method for individual and group of users |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111178417A true CN111178417A (en) | 2020-05-19 |
Family
ID=70652123
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911337331.3A Pending CN111178417A (en) | 2019-12-23 | 2019-12-23 | Energy accurate load prediction method for individual and group of users |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111178417A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115409292A (en) * | 2022-10-31 | 2022-11-29 | 广东电网有限责任公司佛山供电局 | Short-term load prediction method for power system and related device |
Citations (11)
| 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 |
| CN105844371A (en) * | 2016-05-19 | 2016-08-10 | 北京中电普华信息技术有限公司 | Electricity customer short-term load demand forecasting method and device |
| CN106779219A (en) * | 2016-12-19 | 2017-05-31 | 北京中电普华信息技术有限公司 | A kind of electricity demand forecasting method and system |
| CN107563539A (en) * | 2017-07-24 | 2018-01-09 | 佛山市顺德区中山大学研究院 | Short-term and long-medium term power load forecasting method based on machine learning model |
| CN108734355A (en) * | 2018-05-24 | 2018-11-02 | 国网福建省电力有限公司 | A kind of short-term electric load method of parallel prediction and system applied to power quality harnessed synthetically scene |
| CN108876054A (en) * | 2018-07-06 | 2018-11-23 | 国网河南省电力公司郑州供电公司 | Short-Term Load Forecasting Method based on improved adaptive GA-IAGA optimization extreme learning machine |
| CN109376772A (en) * | 2018-09-28 | 2019-02-22 | 武汉华喻燃能工程技术有限公司 | A kind of Combination power load forecasting method based on neural network model |
| KR20190062741A (en) * | 2017-11-29 | 2019-06-07 | 동아대학교 산학협력단 | Electric Power Consumption Analyzing Method Using Deep Learning in Remote Power Control System |
| CN110163659A (en) * | 2019-04-17 | 2019-08-23 | 平安城市建设科技(深圳)有限公司 | Data estimation method, apparatus, equipment and storage medium |
| US20190340467A1 (en) * | 2018-05-06 | 2019-11-07 | Strong Force TX Portfolio 2018, LLC | Facility level transaction-enabling systems and methods for provisioning and resource allocation |
| CN110459040A (en) * | 2019-08-15 | 2019-11-15 | 云南恒协科技有限公司 | A kind of cooperative monitoring system and method |
-
2019
- 2019-12-23 CN CN201911337331.3A patent/CN111178417A/en active Pending
Patent Citations (11)
| 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 |
| CN105844371A (en) * | 2016-05-19 | 2016-08-10 | 北京中电普华信息技术有限公司 | Electricity customer short-term load demand forecasting method and device |
| CN106779219A (en) * | 2016-12-19 | 2017-05-31 | 北京中电普华信息技术有限公司 | A kind of electricity demand forecasting method and system |
| CN107563539A (en) * | 2017-07-24 | 2018-01-09 | 佛山市顺德区中山大学研究院 | Short-term and long-medium term power load forecasting method based on machine learning model |
| KR20190062741A (en) * | 2017-11-29 | 2019-06-07 | 동아대학교 산학협력단 | Electric Power Consumption Analyzing Method Using Deep Learning in Remote Power Control System |
| US20190340467A1 (en) * | 2018-05-06 | 2019-11-07 | Strong Force TX Portfolio 2018, LLC | Facility level transaction-enabling systems and methods for provisioning and resource allocation |
| CN108734355A (en) * | 2018-05-24 | 2018-11-02 | 国网福建省电力有限公司 | A kind of short-term electric load method of parallel prediction and system applied to power quality harnessed synthetically scene |
| CN108876054A (en) * | 2018-07-06 | 2018-11-23 | 国网河南省电力公司郑州供电公司 | Short-Term Load Forecasting Method based on improved adaptive GA-IAGA optimization extreme learning machine |
| CN109376772A (en) * | 2018-09-28 | 2019-02-22 | 武汉华喻燃能工程技术有限公司 | A kind of Combination power load forecasting method based on neural network model |
| CN110163659A (en) * | 2019-04-17 | 2019-08-23 | 平安城市建设科技(深圳)有限公司 | Data estimation method, apparatus, equipment and storage medium |
| CN110459040A (en) * | 2019-08-15 | 2019-11-15 | 云南恒协科技有限公司 | A kind of cooperative monitoring system and method |
Non-Patent Citations (2)
| Title |
|---|
| 徐西蒙;杨任农;符颖;赵雨;: "基于ELM__AdaBoost强预测器的空战目标威胁评估" * |
| 马立新;尹晶晶;郑晓栋;栾健;: "智能用电环境下电力负荷预测方法的研究" * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115409292A (en) * | 2022-10-31 | 2022-11-29 | 广东电网有限责任公司佛山供电局 | Short-term load prediction method for power system and related device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106022528B (en) | A kind of photovoltaic plant short term power prediction technique based on density peaks hierarchical clustering | |
| Rodríguez Fernández et al. | Using the Big Data generated by the Smart Home to improve energy efficiency management | |
| CN105787588B (en) | Dynamic peak-valley time-of-use electricity price method for improving new energy consumption capability | |
| Urgun et al. | Importance sampling using multilabel radial basis classification for composite power system reliability evaluation | |
| CN115796393B (en) | Energy management optimization method, system and storage medium based on multi-energy interaction | |
| CN103390200A (en) | Photovoltaic power station electricity generation output power forecasting method based on similar days | |
| CN107038535A (en) | A kind of intelligent micro-grid building load electricity consumption dispatching method for improving gravitation search | |
| Ye et al. | Combined approach for short-term wind power forecasting based on wave division and Seq2Seq model using deep learning | |
| CN112149890A (en) | Comprehensive energy load prediction method and system based on user energy label | |
| CN111461921B (en) | Load modeling typical user database updating method based on machine learning | |
| CN108062720A (en) | A kind of load forecasting method based on similar day selection and random forests algorithm | |
| CN112994115A (en) | New energy capacity configuration method based on WGAN scene simulation and time sequence production simulation | |
| CN115759467A (en) | Time-division integrated learning photovoltaic prediction method for error correction | |
| CN109829571A (en) | A kind of short-term electricity demand forecasting method of user based on multilist fused data | |
| CN109615139A (en) | A kind of long-term electricity demand forecasting method in the resident based on cultural genetic algorithm | |
| CN116826710A (en) | Peak clipping strategy recommendation method and device based on load prediction and storage medium | |
| CN108694475B (en) | Short-time-scale photovoltaic cell power generation capacity prediction method based on hybrid model | |
| CN111178417A (en) | Energy accurate load prediction method for individual and group of users | |
| CN105279582B (en) | Super short-period wind power prediction technique based on dynamic correlation feature | |
| CN114021837A (en) | Regional power consumption prediction method based on hybrid machine learning and spatial address matching | |
| CN116662860A (en) | User portrait and classification method based on energy big data | |
| CN116011871A (en) | Method for quantifying space update probability of traditional villages based on space update sensitivity | |
| CN101155369A (en) | Power-saving system and method for mobile device | |
| Nadeem et al. | Short term load forecasting on PRECON dataset | |
| Vallejo-Correa et al. | Evaluation of heuristic techniques for solving the short-term hydrothermal scheduling based on key performance indicators (kpis) |
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 |
Application publication date: 20200519 |
|
| RJ01 | Rejection of invention patent application after publication |