CN112862143A - Load and price prediction method - Google Patents
Load and price prediction method Download PDFInfo
- Publication number
- CN112862143A CN112862143A CN201911186559.7A CN201911186559A CN112862143A CN 112862143 A CN112862143 A CN 112862143A CN 201911186559 A CN201911186559 A CN 201911186559A CN 112862143 A CN112862143 A CN 112862143A
- Authority
- CN
- China
- Prior art keywords
- prediction
- load
- price
- forecasting
- interval
- 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 35
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 17
- 238000012216 screening Methods 0.000 claims abstract description 11
- 239000000446 fuel Substances 0.000 claims abstract description 4
- 238000004519 manufacturing process Methods 0.000 claims abstract description 4
- 238000013277 forecasting method Methods 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000010801 machine learning Methods 0.000 claims description 4
- 238000007637 random forest analysis Methods 0.000 claims description 4
- 238000013135 deep learning Methods 0.000 claims description 3
- 238000012417 linear regression Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 238000012935 Averaging Methods 0.000 claims 1
- 230000005611 electricity Effects 0.000 abstract description 4
- 230000006872 improvement Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000000903 blocking effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012360 testing method 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
- G06Q10/04—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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0206—Price or cost determination based on market factors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The invention provides a load and price prediction method, which comprises the following steps: the method comprises the following steps: constructing a historical data set; step two: constructing a second training set; step three: and constructing a prediction model for prediction. The historical data set includes historical price, load, temperature data, humidity, fuel price, production scale, etc. And constructing a second training set, namely sequentially performing first-layer screening according to the type of the day to be predicted and performing second-layer screening according to the average temperature of the day to be predicted. The step of predicting by the established prediction model comprises point prediction and interval prediction, and K day data sets which enable the result error of the prediction method to be minimum are selected firstly, and then prediction is carried out according to a prediction algorithm. The parameter K selected by the historical data considering the time influence in the prediction method provided by the invention enables the prediction to be more accurate, is simultaneously suitable for predicting the electricity price and the load, can perform point prediction and interval prediction, and is more flexible to use.
Description
Technical Field
The invention belongs to the technical field of load and price prediction in the energy field, and particularly relates to a load and price prediction method.
Background
The prediction of load and price is always a key problem in the energy field, and plays an important role in later operation optimization, energy structure adjustment and national economy adjustment. Generally, it is common to predict the load or price 24 hours in the future, and many researches have been made on price and load, but most of them are dedicated to single price prediction and load prediction, such as: load prediction, such as a short-term load prediction method based on kernel principal component analysis and random forests and a power load prediction model based on big data technology; price prediction, such as short-term electricity price prediction based on BP neural network and Markov chain, and design method of dynamic peak electricity price mechanism. The above all belong to a single prediction model, which lacks uniformity and universality, and in fact, both load and price are affected by day type and temperature, and under different market environments, the load and price are affected by many other non-deterministic factors which are difficult to detect and quantify, such as uncertainty of output obtained from other units, influence of information blocking, influence of power grid constraints, and the like, and the non-deterministic factors are difficult to be incorporated into the prediction model, so that accurate point prediction is much more complicated than imagination, therefore, sometimes interval prediction also becomes a direction which can be executed in a falling place, and the interval prediction gives a range of future variation fluctuation.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a prediction method which can be suitable for load prediction and price prediction, and can perform point prediction and interval prediction.
The invention provides a load and price forecasting method, which comprises the following steps:
the method comprises the following steps: constructing a historical data set;
step two: constructing a second training set;
step three: and constructing a prediction model for prediction.
Preferably, the historical data set includes historical price, load, and temperature data.
Preferably, the historical data set further comprises humidity, fuel price, production scale.
Preferably, 60-70% of the historical data set is used as an initial training set.
Preferably, the constructing of the second training set requires two layers of screening, which are:
(1) performing first-layer screening according to the day type of the day to be predicted;
(2) and carrying out second-layer screening according to the average temperature of the day to be predicted.
The preferred embodiment is that the building of the prediction model for prediction specifically includes two situations, namely point prediction and interval prediction, which are respectively:
(1) situation of point prediction
Selecting a K-day data set which enables the error of the result of the prediction method to be minimum in the second training set, and then performing point prediction by using a point prediction algorithm;
(2) situation of interval prediction
And selecting a K-day data set which enables the error of the result of the prediction algorithm to be minimum in the second training set, and then performing interval prediction by using an interval prediction algorithm.
In a preferred embodiment, the point prediction algorithm includes a time series class, a neural network class, a random forest or a linear regression.
In a preferred embodiment, the point prediction algorithm further comprises a method of using a weighted average of the K-day data set as the predicted value.
Preferably, the prediction model is constructed for prediction, and the interval prediction algorithm includes machine learning, deep learning, a probabilistic prediction method, a mixed structure interval method or support vector quantile regression.
In a preferred embodiment, the interval prediction algorithm further includes a method of using interval statistics of the K-day data set as a prediction interval.
The invention has the advantages that: the load and price prediction method provided by the invention has the advantages that the parameter K selected by the historical data considering the time influence is more accurate in prediction, and meanwhile, the method is suitable for prediction of electricity price and load, can be used for point prediction and interval prediction, and is more flexible to use.
Drawings
Fig. 1 is a schematic flow chart of a load and price prediction method provided by the present invention.
Detailed Description
In the following, only certain exemplary embodiments are briefly described. As those skilled in the art can appreciate, the described embodiments can be modified in various different ways, without departing from the spirit or scope of the present disclosure. Accordingly, the drawings and description are to be regarded as illustrative in nature, and not as restrictive.
In the description of the present disclosure, it is to be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "straight", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", and the like, indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience of description and simplicity of description, but do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed in a particular orientation, and be operated, and therefore should not be considered as limiting the present disclosure. Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", may explicitly or implicitly include one or more of the described features. In the description of the present disclosure, "a plurality" means two or more unless specifically limited otherwise.
Throughout the description of the present disclosure, it is to be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection, either mechanically, electrically, or otherwise in communication with one another; either directly or indirectly through intervening media, either internally or in any other relationship. The specific meaning of the above terms in the present disclosure can be understood by those of ordinary skill in the art as appropriate.
In the present disclosure, unless expressly stated or limited otherwise, the first feature "on" or "under" the second feature may comprise the first and second features being in direct contact, or may comprise the first and second features being in contact, not directly, but via another feature in between. Also, the first feature being "on," "above" and "over" the second feature includes the first feature being directly on and obliquely above the second feature, or merely indicating that the first feature is at a higher level than the second feature. A first feature being "under," "below," and "beneath" a second feature includes the first feature being directly above and obliquely above the second feature, or simply meaning that the first feature is at a lesser level than the second feature.
The following disclosure provides many different embodiments or examples for implementing different features of the disclosure. To simplify the disclosure of the present disclosure, specific example components and arrangements are described below. Of course, they are merely examples and are not intended to limit the present disclosure. Moreover, the present disclosure may repeat reference numerals and/or reference letters in the various examples, which have been repeated for purposes of simplicity and clarity and do not in themselves dictate a relationship between the various embodiments and/or arrangements discussed. In addition, the present disclosure provides examples of various specific processes and materials, but one of ordinary skill in the art may recognize applications of other processes and/or use of other materials.
The preferred embodiments of the present disclosure will be described in conjunction with the appended drawings, it being understood that the preferred embodiments described herein are merely for purposes of illustrating and explaining the present disclosure and are not intended to limit the present disclosure
The invention is realized by the following technical scheme:
as shown in fig. 1, the present invention provides a load and price forecasting method, which specifically includes the following steps:
the method comprises the following steps: construction of a historical data set:
the historical data set may contain only historical price/load, temperature, but also other influencing factors such as humidity, fuel price, production scale, etc. Typically, 60% -70% (preferably 67%, this ratio is selected according to actual needs and is not specifically restricted) of the historical data set can be used as the initial training set/test set (also referred to as the first training set).
Step two: and (3) constructing a second training set:
(1) first, a first-layer screening is performed according to the type of the day to be predicted, for example, if the day to be predicted is wednesday, only data of wednesday is selected.
(2) And then, performing a second-layer screening according to the average temperature of the day to be predicted, for example, if the average temperature of the day to be predicted is 20 ℃, selecting data with the average temperature of 15 ℃ to 25 ℃ on the basis of the first-layer screening (the temperature fluctuation range can be 5 ℃ above and below, and the temperature is generally 5 ℃ according to the obvious perception of the human body on the temperature) as a second training set.
Step three: and (5) constructing a prediction model.
(1) Situation of point prediction
According to the influence of data on the result, namely the influence of data closer to the day to be predicted on the result is larger, but the nearest days can reach the best algorithm precision, a parameter K is set, a K-day data set which enables the result error of the prediction method to be the smallest is selected in the second training set, the prediction algorithm is selected in various ways, and the prediction algorithm can be tried by using a time series type, a neural network type, other machine learning algorithms such as random forest, linear regression and the like, even a naive algorithm can be used, for example, the weighted average of the K-day data set can be directly used as a predicted value, and the point prediction result is finally obtained, for example: the predicted load value at a certain time of a day is 2450 kilowatt-hour. The choice of K interacts with the prediction method and needs to be chosen such that the resulting error of the prediction method is minimal.
(2) Situation of interval prediction
Similarly, a K-day data set which enables the result error of the prediction algorithm to be minimum is selected, the selection of the prediction algorithm is diversified at this time, machine learning/deep learning algorithms in point prediction can be tried, besides, a probabilistic prediction method, a mixed structure interval method, support vector quantile regression, a naive algorithm similar to point prediction, for example, a naive algorithm which takes interval statistics of the K-day data set as a prediction interval can be tried, and finally an interval prediction result is obtained, such as: the prediction interval for load at a certain time of day at a 90% confidence level is [2263, 2637 ].
The invention is based on a load and price forecasting framework of various optional algorithms, is suitable for forecasting the load and price in the energy field at the same time, and can carry out point forecasting and interval forecasting.
The above description is only exemplary of the present disclosure and should not be taken as limiting the disclosure, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present disclosure should be included in the scope of protection of the present disclosure.
Finally, it should be noted that: although the present disclosure 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 disclosure. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.
Claims (10)
1. A load and price forecasting method is characterized by comprising the following steps:
the method comprises the following steps: constructing a historical data set;
step two: constructing a second training set;
step three: and constructing a prediction model for prediction.
2. The load, price prediction method of claim 1, wherein the historical data set comprises historical price, load, and temperature data.
3. The load, price prediction method of claim 2, wherein the historical data set further comprises humidity, fuel price, and production scale.
4. A load and price forecasting method according to claim 2 or 3, characterized in that 60-70% of the historical data set is used as an initial training set.
5. The method of claim 1, wherein the second training set is constructed by two layers of screening, which are:
(1) performing first-layer screening according to the day type of the day to be predicted;
(2) and carrying out second-layer screening according to the average temperature of the day to be predicted.
6. The load and price prediction method according to claim 1, wherein the building of the prediction model for prediction specifically includes point prediction and interval prediction, which are respectively:
(1) situation of point prediction
Selecting a K-day data set which enables the error of the result of the prediction method to be minimum in the second training set, and then performing point prediction by using a point prediction algorithm;
(2) situation of interval prediction
And selecting a K-day data set which enables the error of the result of the prediction algorithm to be minimum in the second training set, and then performing interval prediction by using an interval prediction algorithm.
7. The load and price forecasting method of claim 6, wherein the point forecasting algorithm comprises a time series class, a neural network class, a random forest or a linear regression.
8. The load, price forecasting method of claim 7, wherein the point forecasting algorithm further comprises a method of weighted averaging of K-day data sets as forecasted values.
9. The load and price forecasting method according to claim 6, wherein the forecasting model is constructed for forecasting, and the interval forecasting algorithm comprises machine learning, deep learning, probabilistic forecasting, mixed construction interval or support vector quantile regression.
10. The load and price forecasting method of claim 9, wherein the interval forecasting algorithm further comprises a method of taking interval statistics of K-day data sets as forecasting intervals.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911186559.7A CN112862143A (en) | 2019-11-28 | 2019-11-28 | Load and price prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911186559.7A CN112862143A (en) | 2019-11-28 | 2019-11-28 | Load and price prediction method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112862143A true CN112862143A (en) | 2021-05-28 |
Family
ID=75985199
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911186559.7A Pending CN112862143A (en) | 2019-11-28 | 2019-11-28 | Load and price prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112862143A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102016012537A1 (en) * | 2015-12-08 | 2017-06-08 | Adobe Systems Incorporated | Weighting attributes for media content-based recommendations |
| CN108280552A (en) * | 2018-02-24 | 2018-07-13 | 合肥工业大学 | Methods of electric load forecasting and system, storage medium based on deep learning |
| CN108647824A (en) * | 2018-05-10 | 2018-10-12 | 广东工业大学 | A kind of electricity price ultra-short term prediction method |
| CN108921322A (en) * | 2018-05-25 | 2018-11-30 | 广东电网有限责任公司电力调度控制中心 | A kind of ultra-short term node load prediction technique and system based on Spot Price |
| CN109670141A (en) * | 2018-11-30 | 2019-04-23 | 网易传媒科技（北京）有限公司 | Prediction technique, system, medium and electronic equipment |
| CN110135630A (en) * | 2019-04-25 | 2019-08-16 | 武汉数澎科技有限公司 | The short term needing forecasting method with multi-step optimization is returned based on random forest |
| CN110232483A (en) * | 2019-06-18 | 2019-09-13 | 国网河北省电力有限公司经济技术研究院 | Deep learning load forecasting method, device and terminal device |
| WO2019217323A1 (en) * | 2018-05-06 | 2019-11-14 | Strong Force TX Portfolio 2018, LLC | Methods and systems for improving machines and systems that automate execution of distributed ledger and other transactions in spot and forward markets for energy, compute, storage and other resources |
-
2019
- 2019-11-28 CN CN201911186559.7A patent/CN112862143A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102016012537A1 (en) * | 2015-12-08 | 2017-06-08 | Adobe Systems Incorporated | Weighting attributes for media content-based recommendations |
| CN108280552A (en) * | 2018-02-24 | 2018-07-13 | 合肥工业大学 | Methods of electric load forecasting and system, storage medium based on deep learning |
| WO2019217323A1 (en) * | 2018-05-06 | 2019-11-14 | Strong Force TX Portfolio 2018, LLC | Methods and systems for improving machines and systems that automate execution of distributed ledger and other transactions in spot and forward markets for energy, compute, storage and other resources |
| CN108647824A (en) * | 2018-05-10 | 2018-10-12 | 广东工业大学 | A kind of electricity price ultra-short term prediction method |
| CN108921322A (en) * | 2018-05-25 | 2018-11-30 | 广东电网有限责任公司电力调度控制中心 | A kind of ultra-short term node load prediction technique and system based on Spot Price |
| CN109670141A (en) * | 2018-11-30 | 2019-04-23 | 网易传媒科技（北京）有限公司 | Prediction technique, system, medium and electronic equipment |
| CN110135630A (en) * | 2019-04-25 | 2019-08-16 | 武汉数澎科技有限公司 | The short term needing forecasting method with multi-step optimization is returned based on random forest |
| CN110232483A (en) * | 2019-06-18 | 2019-09-13 | 国网河北省电力有限公司经济技术研究院 | Deep learning load forecasting method, device and terminal device |
Non-Patent Citations (1)
| Title |
|---|
| 李干琼: "《SV因子分析框架下的农产品市场短期预测》", 中国经济出版社, pages: 26 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Abuella et al. | Solar power forecasting using artificial neural networks | |
| US9778639B2 (en) | Systems and methods for adaptively updating equipment models | |
| EP1907996B1 (en) | Load prediction based on-line and off-line training of neural networks | |
| US9287713B2 (en) | Topology identification in distribution network with limited measurements | |
| US11107094B2 (en) | Prediction system and prediction method | |
| Von Krannichfeldt et al. | Online ensemble learning for load forecasting | |
| McCarthy et al. | Advanced process control of irrigation: the current state and an analysis to aid future development | |
| WO2019056499A1 (en) | Prediction model training method, data monitoring method, apparatuses, device and medium | |
| US11585549B1 (en) | Thermal modeling technology | |
| Park et al. | A framework for baseline load estimation in demand response: Data mining approach | |
| CN111695736B (en) | Photovoltaic power generation short-term power prediction method based on multi-model fusion | |
| US11815863B2 (en) | Predictive modeling and control for water resource infrastructure | |
| CN109035067A (en) | Building energy consumption processing method and processing device based on RF and ARMA algorithm | |
| EP3905424A1 (en) | Data processing device, data processing method and computer program | |
| CN109118012A (en) | A kind of industrial dynamics various dimensions energy consumption cost prediction technique, system, storage medium and terminal | |
| CN112488192A (en) | Regional power grid short-term load prediction method | |
| CN114595873A (en) | Grey-correlation-based DA-LSTM short-term power load prediction method | |
| CN112862142A (en) | Load and price prediction and correction method | |
| CN109523077A (en) | A kind of wind power forecasting method | |
| CN112862143A (en) | Load and price prediction method | |
| CN109214458A (en) | A kind of city load quantization method based on historical data | |
| KR102266494B1 (en) | System and Method of Deep Learning-based Customer Classification for Demand Respond | |
| CN114372895A (en) | Method for analyzing power load curve based on kmeans algorithm | |
| EP4170439B1 (en) | Optimized hvac control using domain knowledge combined with deep reinforcement learning (drl) | |
| CN117791626B (en) | Intelligent comprehensive power box power supply optimization method |
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 |