CN115619450A - Virtual power plant real-time electricity price prediction system - Google Patents
Virtual power plant real-time electricity price prediction system Download PDFInfo
- Publication number
- CN115619450A CN115619450A CN202211396140.6A CN202211396140A CN115619450A CN 115619450 A CN115619450 A CN 115619450A CN 202211396140 A CN202211396140 A CN 202211396140A CN 115619450 A CN115619450 A CN 115619450A
- Authority
- CN
- China
- Prior art keywords
- data
- real
- time
- electricity price
- subsystem
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000005611 electricity Effects 0.000 title claims abstract description 57
- 238000000034 method Methods 0.000 claims abstract description 22
- 238000010801 machine learning Methods 0.000 claims abstract description 17
- 230000008569 process Effects 0.000 claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 4
- 238000012937 correction Methods 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 abstract description 5
- 238000010248 power generation Methods 0.000 description 10
- 238000012549 training Methods 0.000 description 4
- 230000002776 aggregation Effects 0.000 description 3
- 238000004220 aggregation Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000004146 energy storage Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000006116 polymerization reaction Methods 0.000 description 2
- 230000001932 seasonal effect Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000002146 bilateral effect Effects 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009897 systematic effect Effects 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
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- General Physics & Mathematics (AREA)
- Finance (AREA)
- Economics (AREA)
- Software Systems (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Tourism & Hospitality (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Game Theory and Decision Science (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a real-time electricity price prediction system of a virtual power plant, which comprises an event-driven information subsystem and a real-time prediction subsystem, wherein the event-driven information subsystem comprises a real-time data processor and is used for downloading historical electricity price data in real time, processing the historical electricity price data and then sending the historical electricity price data to the real-time prediction subsystem; and the real-time prediction subsystem receives the data information of the event-driven information subsystem, processes the collected data, inputs the processed data into a machine learning model to predict real-time electricity price or power demand, and feeds the predicted real-time electricity price or power demand back to the event-driven information subsystem. The system realizes low-price high-price selling, profit and energy efficiency optimization of the virtual power plant through accurate prediction of market electricity price.
Description
Technical Field
The application relates to the technical field of power systems, in particular to a real-time electricity price prediction system of a virtual power plant.
Background
In recent years, distributed power sources that generate electricity by using renewable energy sources have been developed rapidly, however, the large-scale development of the distributed power sources is limited by the characteristics of small capacity, large quantity, distributed access and intermittent output. Renewable energy is polymerized in a virtual power plant mode, and the energy management system controls each polymerization unit, so that the coordinated optimization operation of the polymerization units can be realized, the stability of the virtual power plant and the competitiveness of the virtual power plant participating in the power market are improved, and the benefit of scale economy is obtained. The electric power markets which the virtual power plant can participate in include a day-ahead market, a real-time market, a bilateral contract market, an auxiliary service market and the like, the participation in the balance market can help the virtual power plant to stabilize the fluctuation of renewable energy sources, reduce the risk of inaccurate output prediction of the renewable energy sources, and obtain greater economic benefits.
At present, the application of a real-time electricity price bidding prediction model in the open power market is not popularized yet. The power dispatching system attached to the power grid mostly adopts an experience method of manual participation to adjust the matching of power supply and demand.
The electric power market publicization is the trend of power grid management of each province and city, wherein the virtual power plant real-time power price prediction system integrates power grid power price bidding prediction of renewable energy sources. Through expansion, the method can also be applied to prediction of real-time electric power market price and energy demand of the smart power grid or the micro power grid.
The virtual power plant real-time electricity price prediction system integration process needs to have deep knowledge of the open power market. Foreign, nextkaffwerke in germany has established a virtual plant solution (SaaS) based on the aggregation of power generation resources. At present, the domestic public power market is not established yet and is basically in the early-stage pilot study stage. Commercial companies in this industry lack complete testing precedent, platform and technology accumulation.
Nextkaffwerke, germany, offers a systematic solution for "software as a service" (SaaS). The company provides an optimization scheme for users according to the information of power generation, energy storage and load equipment provided by the users. However, the technology of the company is huge and expensive, and the company is difficult to be suitable for the market of China. The complete solution can only provide consultation for large-scale power generation enterprises or power grid systems. However, due to its highly integrated characteristic, it is unable to provide decision basis for decentralized power generation, energy storage, and load scheduling.
Disclosure of Invention
Aiming at the problems, the invention provides a real-time electricity price prediction system of a virtual power plant, which realizes the prediction of real-time bidding of a power grid by combining a machine learning algorithm at the front, namely a decision tree algorithm for time sequence prediction and natural language processing.
The invention discloses a real-time electricity price forecasting system of a virtual power plant, which comprises an event-driven information subsystem and a real-time forecasting subsystem, wherein the event-driven information subsystem comprises a real-time data processor and is used for downloading historical electricity price data in real time, processing the historical electricity price data and then sending the historical electricity price data to the real-time forecasting subsystem; and the real-time prediction subsystem receives the data information of the event-driven information subsystem, processes the collected data, inputs the processed data into a machine learning model to predict real-time electricity price or power demand, and feeds the predicted real-time electricity price or power demand back to the event-driven information subsystem.
The invention further adopts the technical scheme that: the historical electricity price data comprises historical electricity prices issued by energy market operators, weather historical data, real-time weather forecast data and energy historical price data.
The further technical scheme of the invention is as follows: the real-time prediction subsystem processes the collected data, and specifically comprises the following steps: tagging non-numeric data, numeric data normalization, numeric data regularization, backfilling missing data, and alignment of multiple data streams.
The further technical scheme of the invention is as follows: the machine learning model extracts the processed data features, and the data features specifically include: electricity price or power demand characteristics, weather characteristics, energy price characteristics, and time characteristics.
The further technical scheme of the invention is as follows: the machine learning model also comprises the following steps of screening the extracted data characteristics: the XGboost algorithm is used for carrying out regression operation on the characteristics and the results of the data so as to establish the relation between the characteristics and the results.
The further technical scheme of the invention is as follows: the machine learning model further comprises the step of carrying out weight correction on the model by using the collected data, and predicting the electricity price 24 hours in the future after the correction is completed.
The invention provides a real-time electricity price forecasting system of a virtual power plant, which is suitable for forecasting the electricity price of a highly marketized public power market so as to achieve the aim of maximizing the energy efficiency of power grid management. The invention has the beneficial effects that: the problems of aggregation and management of power generation resources of a supply end are solved, and low-price high-price selling, profit and energy efficiency optimization of a virtual power plant are realized through accurate prediction of market power price; the virtual power plant integrating the hardware system, and the micro-grid system can allow users of enterprises, residents and the like to participate in electric power market transactions.
Drawings
FIG. 1 is a schematic structural diagram of a real-time electricity price prediction system of a virtual power plant in an embodiment of the invention;
fig. 2 is a schematic diagram of an implementation method of a real-time prediction subsystem according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Before discussing exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although a flowchart may describe the steps as a sequential process, many of the steps can be performed in parallel, concurrently, or simultaneously. In addition, the order of the steps may be rearranged. The process may be terminated when its operations are completed, but could have additional steps not included in the figure. The processes may correspond to methods, functions, procedures, subroutines, and the like.
The embodiment of the invention aims at a real-time electricity price forecasting system of a virtual power plant, and provides the following embodiments:
referring to fig. 1, a real-time electricity price prediction system 100 of a virtual power plant in embodiment 1 of the present invention includes an event-driven information subsystem 101 and a real-time prediction subsystem 102, where the event-driven information subsystem 101 includes a real-time data processor, and is configured to download historical electricity price data in real time, process the historical electricity price data, and send the historical electricity price data to the real-time prediction subsystem 102, and the event-driven information subsystem 101 further receives prediction completion information fed back by the real-time prediction subsystem 102, and sends the prediction completion information to a downstream user for sharing (data shared by a system on the power generation side in fig. 1); the real-time prediction subsystem 102 receives the data information of the event-driven information subsystem 101, processes the collected data, inputs the processed data into a machine learning model to predict real-time electricity prices or power demands, and feeds back the predicted real-time electricity prices or power demands to the event-driven information subsystem 101.
Referring to fig. 1-2, the historical electricity price data collected by the power generation side includes historical electricity prices issued by energy market operators, weather historical data, real-time weather forecast data, and energy historical price data.
Referring to fig. 1-2, the real-time prediction subsystem 102 processes the historical electricity price data collected by the power generation side, and specifically includes: tagging non-numeric data, numeric data normalization, numeric data regularization, backfilling missing data, and alignment of multiple data streams.
Referring to fig. 1-2, the machine learning model extracts processed data features, where the data features specifically include: electricity price or power demand characteristics, weather characteristics, energy price characteristics, and time characteristics.
Referring to fig. 1-2, the machine learning model further includes a step of screening the extracted data features, specifically: the XGboost algorithm is used for carrying out regression operation on the characteristics and the results of the data so as to establish the relation between the characteristics and the results.
Referring to fig. 1-2, the machine learning model further includes a weight correction for the model using the collected data, and predicting a future electricity rate for 24 hours after the correction is completed.
In a specific implementation, the real-time forecasting subsystem 102 is combined with the event-driven information subsystem 101 to achieve automatic reading of data related to the model and forecasting of electricity prices and power demands of the real-time power market. The event driven information subsystem 101 is responsible for awaiting the arrival of the information on the generating side. When the power generation side information indicates that there is a new data update, the event driven information subsystem 101 downloads the updated data according to the content indicated by this information. Next, the event-driven subsystem 101 processes the newly downloaded data and sends the processed data to the real-time prediction subsystem 102 for prediction. After the prediction is completed, the real-time prediction subsystem 102 feeds back the prediction completion information to the event-driven subsystem 101, and the event-driven subsystem 101 sends the prediction completion information to the downstream module.
More specifically, the information received by the event-driven subsystem 101 includes an update notification of data such as historical electricity prices, weather conditions, air temperatures, seasonal changes, work festivals and holidays, and the like. The data downloaded by the event driven subsystem 101 includes historical electricity prices, weather conditions, air temperatures, seasonal changes, work holidays, and the like. The event-driven subsystem 101 performs gap filling and tag arrangement operations on the downloaded data updates. When the real-time prediction subsystem 102 completes the prediction task, the event-driven subsystem 101 sends the prediction completion information to the downstream user.
The real-time prediction subsystem 102 first receives a data update notification (message) from the event driven subsystem 101, which triggers the data loading function of the real-time prediction subsystem 102. After the data is read, the real-time prediction subsystem 102 normalizes, denoises, encodes, matrixs, and inputs the data to a machine learning model for weight correction (training) and result prediction. In the training process of the model, specifically, the XGboost algorithm is used by the machine learning model to perform Regression operation (Regression) on the characteristics and the result of the input data so as to establish the connection between the characteristics and the result and achieve the purpose of 'supervised learning'. Through training of the machine learning model by historical data, the output result of the real-time prediction subsystem 102 is a predicted value of the electricity price or the power demand for half an hour to twenty-four hours in the future. Due to the real-time nature of the data, the real-time prediction subsystem 102 always performs training and weight adjustment based on the newly loaded data. After the model weight is updated, the real-time prediction subsystem 102 predicts the electricity price of the future 24 hours as an output, and stores the predicted data. At this time, the event driven subsystem 101 sends a notification (prediction completion information) to the downstream user, and the downstream user decides whether to read the prediction data. Therefore, the system completes the automatic process of reading, processing and sending information/data. Downstream users comprise decision programs or decision makers such as a micro-grid and a virtual power plant, and due to the fact that the decision makers need to carry out power supply and demand allocation on the micro-grid or carry out power supply bidding outside, the fluctuation situation of the future power price provides necessary guidance for decision making.
The virtual power plant real-time electricity price prediction system provided by the embodiment is integrated, and is suitable for predicting the electricity price of a highly marketized public power market so as to achieve the purpose of maximizing the energy efficiency of power grid management. The invention has the beneficial effects that: the problems of aggregation and management of power generation resources of a supply end are solved, and low-price high-price selling, profit and energy efficiency optimization of a virtual power plant are realized through accurate prediction of market power price; the virtual power plant integrated with the hardware system, and the micro-grid system can allow users such as enterprises and residents to participate in electric power market transactions.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in some detail by the above embodiments, the invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the invention, and the scope of the invention is determined by the scope of the appended claims.
Claims (6)
1. The virtual power plant real-time electricity price prediction system is characterized by comprising an event-driven information subsystem and a real-time prediction subsystem, wherein the event-driven information subsystem comprises a real-time data processor and is used for downloading historical electricity price data in real time, processing the historical electricity price data and then sending the historical electricity price data to the real-time prediction subsystem, receiving prediction completion information fed back by the real-time prediction subsystem by the event-driven information subsystem and sending the prediction completion information to downstream users for sharing; and the real-time prediction subsystem receives the data information of the event-driven information subsystem, processes the collected data, inputs the processed data into a machine learning model to predict real-time electricity price or power demand, and feeds the predicted real-time electricity price or power demand back to the event-driven information subsystem.
2. The real-time power price prediction system of the virtual power plant as claimed in claim 1, wherein the historical power price data comprises historical power price issued by energy market operators, weather historical data, real-time weather prediction data and energy historical price data.
3. The virtual power plant real-time electricity price prediction system of claim 1, wherein the real-time prediction subsystem processes the collected data, specifically comprising: tagging non-numeric data, numeric data normalization, numeric data regularization, backfilling missing data, and alignment of multiple data streams.
4. The real-time electricity price prediction system of a virtual power plant according to claim 1, characterized in that the machine learning model extracts processed data features, the data features specifically comprising: electricity price or power demand characteristics, weather characteristics, energy price characteristics, and time characteristics.
5. The virtual power plant real-time electricity price prediction system of claim 1, wherein the machine learning model further comprises screening the extracted data features, specifically: the XGboost algorithm is used for carrying out regression operation on the characteristics and the results of the data so as to establish the relation between the characteristics and the results.
6. The real-time power plant price forecasting system of claim 1, wherein the machine learning model further comprises using the collected data to perform weight correction on the model, and predicting the electricity price 24 hours in the future after the correction is completed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211396140.6A CN115619450A (en) | 2022-11-09 | 2022-11-09 | Virtual power plant real-time electricity price prediction system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211396140.6A CN115619450A (en) | 2022-11-09 | 2022-11-09 | Virtual power plant real-time electricity price prediction system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115619450A true CN115619450A (en) | 2023-01-17 |
Family
ID=84879449
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211396140.6A Pending CN115619450A (en) | 2022-11-09 | 2022-11-09 | Virtual power plant real-time electricity price prediction system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115619450A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116342234A (en) * | 2023-05-26 | 2023-06-27 | 山东纵横易购产业互联网有限公司 | Method for realizing automatic bidding purchasing aiming at goods |
| CN116720885A (en) * | 2023-08-07 | 2023-09-08 | 国网安徽省电力有限公司经济技术研究院 | Distributed virtual power plant control method and system in electric power spot market environment |
-
2022
- 2022-11-09 CN CN202211396140.6A patent/CN115619450A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116342234A (en) * | 2023-05-26 | 2023-06-27 | 山东纵横易购产业互联网有限公司 | Method for realizing automatic bidding purchasing aiming at goods |
| CN116342234B (en) * | 2023-05-26 | 2023-08-29 | 山东纵横易购产业互联网有限公司 | Method for realizing automatic bidding purchasing aiming at goods |
| CN116720885A (en) * | 2023-08-07 | 2023-09-08 | 国网安徽省电力有限公司经济技术研究院 | Distributed virtual power plant control method and system in electric power spot market environment |
| CN116720885B (en) * | 2023-08-07 | 2023-10-20 | 国网安徽省电力有限公司经济技术研究院 | Distributed virtual power plant control method and system in electric power spot market environment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7482167B2 (en) | SYSTEM AND METHOD FOR DYNAMIC ENERGY STORAGE SYSTEM CONTROL - Patent application | |
| US10261536B2 (en) | Systems and methods for optimizing microgrid power generation and management with predictive modeling | |
| CN115619450A (en) | Virtual power plant real-time electricity price prediction system | |
| JP2023129546A (en) | System and method for optimal control of energy storage system | |
| CN105761109A (en) | Smart management system for energy management and electricity transaction in virtual power plant and method for optimizing operation thereof | |
| US20120253532A1 (en) | Systems and methods for forecasting electrical load | |
| US11029652B2 (en) | Systems, methods and devices for managing/controlling energy production and/or energy storage systems | |
| CA3030490A1 (en) | Intelligent energy management system for distributed energy resources and energy storage systems using machine learning | |
| US12015275B2 (en) | Hybrid power plant | |
| CN111105081A (en) | Energy-saving power generation scheduling technology support method, device and equipment based on spot market | |
| US20210055700A1 (en) | Hybrid machine learning and simulation based system for forecasting in electricity systems | |
| Gao et al. | Green Energy Cloud-Taxonomy, Infrastructure, Platform, and Services | |
| Du et al. | Integrating a multi-microgrid system into real-time balancing market: Problem formulation and solution technique | |
| Zandi et al. | An automatic learning framework for smart residential communities | |
| Aragón et al. | Optimization Framework for short-term control of Energy Storage Systems | |
| GB2599054A (en) | Systems and methods for continuous optimization and real-time control of multiple energy assets | |
| CN117674300B (en) | Virtual power plant resource scheduling method and device, terminal equipment and storage medium | |
| Tina et al. | Aggregation Platform to Maximize the Utilization of Distributed Energy Resources' Flexibility | |
| EP4371807A1 (en) | Optimized control of power depots using multi-tenant charging system-of-systems | |
| CN116247738A (en) | Virtual power plant energy management method based on micro-grid dispatching | |
| CN113823075B (en) | Intelligent adjustment method and device for request waiting time delay of concentrator high-efficiency meter reading equipment | |
| Zavadskiy et al. | Involving renewable energy sources in generating process: example of Kazakhstan | |
| WO2024105136A1 (en) | Optimized control of power depots using multi-tenant charging system-of-systems | |
| CN116720696A (en) | Optimization method, terminal and storage medium for park-type virtual power plant group | |
| Guan et al. | Method and Application of Energy Storage Spot Trading Based on Dual Settlement Market Model |
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 |