CN115102202B - Energy storage control method based on rolling type real-time electricity price prediction - Google Patents
Energy storage control method based on rolling type real-time electricity price prediction Download PDFInfo
- Publication number
- CN115102202B CN115102202B CN202210880549.9A CN202210880549A CN115102202B CN 115102202 B CN115102202 B CN 115102202B CN 202210880549 A CN202210880549 A CN 202210880549A CN 115102202 B CN115102202 B CN 115102202B
- Authority
- CN
- China
- Prior art keywords
- real
- electricity price
- time
- prediction
- time electricity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000005611 electricity Effects 0.000 title claims abstract description 145
- 238000005096 rolling process Methods 0.000 title claims abstract description 75
- 238000004146 energy storage Methods 0.000 title claims abstract description 72
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 54
- 238000003066 decision tree Methods 0.000 claims abstract description 44
- 238000011217 control strategy Methods 0.000 claims abstract description 26
- 238000007619 statistical method Methods 0.000 claims abstract description 20
- 238000005457 optimization Methods 0.000 claims abstract description 19
- 239000011664 nicotinic acid Substances 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 239000000126 substance Substances 0.000 claims description 5
- 238000012549 training Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 5
- 238000007599 discharging Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000013461 design Methods 0.000 description 4
- 230000005684 electric field Effects 0.000 description 4
- 238000009826 distribution Methods 0.000 description 2
- 230000003203 everyday effect Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/28—Arrangements for balancing of the load in a network by storage of energy
-
- 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
- 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
- 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—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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/28—Arrangements for balancing of the load in a network by storage of energy
- H02J3/32—Arrangements for balancing of the load in a network by storage of energy using batteries with converting means
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
- H02J3/381—Dispersed generators
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/20—The dispersed energy generation being of renewable origin
- H02J2300/28—The renewable source being wind energy
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2310/00—The network for supplying or distributing electric power characterised by its spatial reach or by the load
- H02J2310/50—The network for supplying or distributing electric power characterised by its spatial reach or by the load for selectively controlling the operation of the loads
- H02J2310/56—The network for supplying or distributing electric power characterised by its spatial reach or by the load for selectively controlling the operation of the loads characterised by the condition upon which the selective controlling is based
- H02J2310/62—The condition being non-electrical, e.g. temperature
- H02J2310/64—The condition being economic, e.g. tariff based load management
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Finance (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Power Engineering (AREA)
- Accounting & Taxation (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Game Theory and Decision Science (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Water Supply & Treatment (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Primary Health Care (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The application provides an energy storage control method and system based on rolling type real-time electricity price prediction, and the method comprises the following steps: calculating a rolling time interval of the real-time electricity price prediction; acquiring actual real-time electricity price data at intervals of a rolling time interval, and performing statistical analysis on the expected difference between the day-ahead price and the real-time electricity price through expected statistical analysis; predicting the real-time electricity price in a future preset time period through a rolling type prediction algorithm, wherein the method comprises the following steps: calculating the system load rate of each moment through a system load rate algorithm model, and performing real-time electricity price prediction through a gradient lifting decision tree based on the system load rate; and correcting the predicted real-time electricity price according to the expected difference value, determining boundary data of energy storage control according to the corrected real-time electricity price, and optimizing a control strategy of the energy storage system. The method can provide more accurate data boundary in a rolling mode for the optimization of the flexible energy storage control strategy, and improve the accuracy of flexible control of the energy storage system.
Description
Technical Field
The application relates to the technical field of energy storage control, in particular to an energy storage control method and system based on rolling type real-time electricity price prediction.
Background
With the increasing demand for electricity, people pay more and more attention to smooth operation of power systems and settlement income of power systems. If the stored energy of the power system cannot be reasonably set, the settlement income and the normal operation of the power system can be directly influenced, and the loss of 'excess profit recovery' can be generated in some cases. For a wind power plant with an energy storage system, the wind power plant has certain adjusting capacity, but how to flexibly control the energy storage system to enable the wind power storage system to jointly obtain the optimal running state is a problem to be solved urgently. The optimization of the flexible control strategy of the energy storage system needs to use real-time electricity price data of a period of time in the future as an important basis, so that the accuracy of real-time electricity price prediction and the rolling update of the real-time electricity price directly influence the quality of the flexible control strategy.
In the related art, when energy storage control is performed based on electricity price prediction, the scheme is adopted to predict real-time electricity price data of a whole day at one time. However, there are two disadvantages to energy storage control by predicting real-time electricity prices in this way: firstly, the prediction accuracy of the real-time electricity price is low, and accurate and effective input data cannot be provided for the flexible energy storage control system, so that the accuracy of the flexible energy storage control strategy is low. Secondly, in some scenarios, a policy error may be caused even if the energy storage battery receives a charge and discharge policy of a reverse operation (for example, at a certain time, the policy signal of receiving charge instead receives the policy signal of discharge or the policy signal of receiving discharge instead receives the policy signal of charge). Both of the above disadvantages may lead to erroneous energy storage control of the electric field provided with the energy storage device.
Therefore, how to reasonably and flexibly control the energy storage system according to accurate real-time electricity prices becomes a problem which needs to be solved urgently at present.
Disclosure of Invention
The present application is directed to solving, at least to some extent, one of the technical problems in the related art.
Therefore, a first object of the present application is to provide an energy storage control method based on rolling type real-time electricity price prediction, which predicts real-time electricity prices in a rolling type prediction manner, uses a price difference between a current price and the real-time electricity prices as an important characteristic of the real-time electricity price prediction, and uses a system load rate as input data of the electricity price prediction, so as to greatly improve the prediction accuracy of the real-time electricity prices, provide a more accurate and reliable price boundary for an energy storage flexible control strategy algorithm, and further improve the reliability and accuracy of a flexible control strategy optimization result.
The second purpose of the application is to provide an energy storage control system based on rolling type real-time electricity price prediction;
a third object of the present application is to propose a non-transitory computer-readable storage medium.
In order to achieve the above object, a first embodiment of the present application is to provide an energy storage control method based on rolling type real-time electricity price prediction, where the method includes the following steps:
calculating a rolling time interval of the real-time electricity price prediction;
acquiring actual real-time electricity price data at intervals of a rolling time interval, and performing statistical analysis on the expected difference between the day-ahead price and the real-time electricity price through expected statistical analysis;
predicting the real-time electricity price in a future preset time period through a rolling type prediction algorithm, wherein the method comprises the following steps: calculating the system load rate of each moment through a preset system load rate algorithm model, and performing real-time electricity price prediction through a gradient lifting decision tree GBDT model based on the system load rate;
and correcting the predicted real-time electricity price according to the expected difference value, determining boundary data of energy storage control according to the corrected real-time electricity price, and optimizing a flexible control strategy of the energy storage system based on the boundary data.
Optionally, in an embodiment of the present application, calculating a rolling time interval of the real-time electricity price prediction includes: acquiring a preset energy station declaration output time interval, a spot market clearing price curve time interval and an ultra-short period power prediction value time interval; calculating a rolling time interval of the real-time electricity rate prediction by the following formula:
M=min{K,L,N}
wherein M is a rolling time interval of real-time electricity price prediction, K is an energy field station declared output time interval, L is a spot market clearing price curve time interval, and N is an ultra-short period power prediction numerical value time interval.
Optionally, in an embodiment of the present application, the system load rate algorithm model is as shown in the following formula:
wherein the content of the first and second substances,is composed oftThe rate of system load at the time of day,is composed oftThe predicted value of the load of the whole network at the moment,is composed oftThe predicted value of the new energy output at the moment,is composed oftThe total capacity of all participating spot units at the moment,tindicating any time of day.
Optionally, in an embodiment of the present application, the real-time electricity price prediction by the GBDT model based on the system load rate includes: establishing decision trees with corresponding number of preset depths according to the system load rate at each moment in the future preset time period; generating a corresponding weak learner according to each decision tree, and acquiring the learned weight of each weak learner; and generating a strong learner algorithm model according to all weak learners and corresponding weights thereof, and predicting the electricity price in real time through the strong learner algorithm model.
Optionally, in an embodiment of the present application, before the generating the strong learner algorithm model, further includes: and optimizing and adjusting parameters of the number of decision trees, the minimum sample number of cotyledons of the decision trees, the maximum depth of the decision trees and the maximum characteristic number ratio in the GBDT model through a bionic optimization parameter adjusting model.
In order to achieve the above object, a second aspect of the present application further provides an energy storage control system based on rolling real-time electricity price prediction, including the following modules:
the calculating module is used for calculating a rolling time interval of the real-time electricity price prediction;
the statistical analysis module is used for acquiring actual real-time electricity price data at intervals of a rolling time interval and carrying out statistical analysis on the expected difference value between the current price and the real-time electricity price through expected statistical analysis;
the prediction module is used for predicting the real-time electricity price in a future preset time period through a rolling type prediction algorithm and comprises the following steps: calculating the system load rate of each moment through a preset system load rate algorithm model, and performing real-time electricity price prediction through a gradient lifting decision tree GBDT model based on the system load rate;
and the control module is used for correcting the predicted real-time electricity price according to the expected difference value, determining boundary data of energy storage control according to the corrected real-time electricity price, and optimizing a flexible control strategy of the energy storage system based on the boundary data.
Optionally, in an embodiment of the present application, the calculation module is specifically configured to: acquiring a preset energy station declaration output time interval, a spot market clearing price curve time interval and an ultra-short period power prediction value time interval; calculating a rolling time interval for the real-time electricity price prediction by the following formula:
M=min{K,L,N}
wherein M is a rolling time interval of real-time electricity price prediction, K is an energy field station declared output time interval, L is a spot market clearing price curve time interval, and N is an ultra-short period power prediction numerical value time interval.
Optionally, in an embodiment of the present application, the prediction module is specifically configured to calculate the system load rate according to the following formula:
wherein the content of the first and second substances,is composed oftThe rate of system load at the time of day,is composed oftThe predicted value of the load of the whole network at the moment,is composed oftThe predicted value of the new energy output at the moment,is composed oftThe total capacity of all participating spot units at the moment,tindicating any time of day.
Optionally, in an embodiment of the present application, the prediction module is further configured to:
establishing decision trees with corresponding number of preset depths according to the system load rate at each moment in the future preset time period;
generating a corresponding weak learner according to each decision tree, and acquiring the learned weight of each weak learner;
and generating a strong learner algorithm model according to all weak learners and corresponding weights thereof, and predicting the electricity price in real time through the strong learner algorithm model.
The technical scheme provided by the embodiment of the application at least has the following beneficial effects: according to the method and the device, the real-time electricity price is predicted in a rolling type prediction mode, and the price difference between the current price and the real-time electricity price is used as important characteristic data for predicting the real-time electricity price, so that the prediction accuracy of the real-time electricity price is greatly improved, and a more accurate and reliable price boundary is provided for an energy storage flexible control strategy algorithm. The prediction mode of the prior art is logically changed from an algorithm, the method better meets the actual operation requirement of an electric field provided with an energy storage device, and can provide more accurate data boundary for the optimization of the flexible energy storage control strategy in a rolling mode, so that the accuracy of the flexible control strategy is improved; in addition, the system load rate algorithm model is added to the algorithm model design, the efficient bionic optimization parameter adjusting model is established to optimize the model parameters, and the accuracy and the efficiency of real-time electricity price prediction are improved. In addition, the real-time electricity price rolling type prediction algorithm model is reasonable and effective in design and high in parameter optimizing speed, and can meet the requirement of the energy storage device in actual operation more fully.
In order to implement the foregoing embodiments, the third aspect of the present application further provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the energy storage control method based on rolling type real-time electricity price prediction in the foregoing embodiments is implemented.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a flowchart of an energy storage control method based on rolling real-time electricity price prediction according to an embodiment of the present application;
FIG. 2 is a logic diagram of a rolling prediction algorithm according to an embodiment of the present application;
fig. 3 is a schematic diagram illustrating a principle of real-time electricity price prediction through a GBDT model and a bionic optimization parameter adjusting model according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of an energy storage control system based on rolling real-time electricity price prediction according to an embodiment of the present application.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
An energy storage control method and system based on rolling type real-time electricity price prediction provided by the embodiment of the application are described below with reference to the accompanying drawings.
Fig. 1 is a flowchart of an energy storage control method based on rolling type real-time electricity price prediction according to an embodiment of the present application, and as shown in fig. 1, the method includes the following steps:
step S101, calculating a rolling time interval of the real-time electricity price prediction.
In an embodiment of the present application, when calculating the rolling time interval of the real-time electricity price prediction, a preset energy field station declared output time interval, a spot market clearing price curve time interval, and an ultra-short term power prediction value time interval may be obtained first. In specific implementation, known data information such as relevant trade rule regulations and historical data of a local power market can be collected, and the declared output time interval K of the energy field station and the output price curve time interval L of the spot market can be obtained. And then, acquiring the ultra-short-term power prediction value time interval N output by the prediction system or the optimization model through the existing power prediction system or the built and trained ultra-short-term power prediction optimization model.
Then, the rolling time interval of the real-time electricity price prediction is calculated by the following formula:
M=min{K,L,N}
wherein M is rolling time interval of real-time electricity price prediction, K is energy field station declared output time interval, L is spot market clearing price curve time interval, and N is ultra-short period power prediction numerical value time interval
And S102, acquiring actual real-time electricity price data at intervals of a rolling time interval, and performing statistical analysis on the expected difference between the day-ahead price and the real-time electricity price through expected statistical analysis.
Specifically, real-time electricity price data is acquired at intervals, namely, at rolling time intervals m, by referring to relevant actual transaction information and the like, and statistical analysis is performed on the expectation of the difference between the day-ahead price and the acquired real-time electricity price through an expectation statistical analysis algorithm. The day-ahead price can be obtained by establishing a day-ahead price prediction model and the like.
Step S103, predicting the real-time electricity price in a future preset time period by a rolling type prediction algorithm, wherein the method comprises the following steps: and calculating the system load rate at each moment through a preset system load rate algorithm model, and performing real-time electricity price prediction through a gradient lifting decision tree GBDT model based on the system load rate.
The Gradient Boosting Decision Tree (GBDT) is an iterative decision tree algorithm, and the algorithm is composed of a plurality of decision trees, and the conclusions of all the trees are accumulated to make a final answer.
Specifically, real-time electricity prices in a future period are predicted through a rolling algorithm model proposed by the present application, wherein logic of the rolling prediction algorithm is as shown in fig. 2, and with reference to an example shown in fig. 2, assuming that a current time is 00.
In one embodiment of the application, the prediction of the real-time electricity price in the future preset time period through the rolling type prediction algorithm comprises two steps, the first step is to calculate the system load rate of each time of the whole day through a pre-established system load rate algorithm model, the second step is to establish a GBDT price prediction algorithm and a bionic optimization parameter adjusting model based on the system load rate, and the real-time electricity price prediction is carried out through a gradient lifting decision tree GBDT model.
Specifically, the system load rate algorithm model in the first step is shown in the following formula:
wherein the content of the first and second substances,is composed oftThe rate of system load at the time of day,is composed oftThe predicted value of the whole network load at the moment,is composed oftThe predicted value of the new energy output at the moment,is composed oftThe total capacity of all participating spot units at the moment,tindicating any time of day.
In specific implementation, the day-ahead disclosure data can be acquired in different ways, and the required data can be read from the day-ahead disclosure data. For example, according to relevant day-ahead disclosure data published by the electric power spot transaction platform, data such as whole network load prediction data, new energy output prediction data, total capacity of units participating in spot goods and the like are read, and the obtained data are input into the system load rate algorithm model, so that the corresponding system load rate at each moment in the whole day can be calculated.
The method comprises the following steps of firstly establishing decision trees with corresponding number and preset depth according to the system load rate at each moment in a preset time period in the future, then generating corresponding weak learners according to each decision tree, acquiring the learned weight of each weak learner, finally generating a strong learner algorithm model according to all the weak learners and the corresponding weights thereof, and carrying out real-time electricity price prediction through the strong learner algorithm model.
Specifically, referring to the schematic diagram of the principle of real-time electricity price prediction through the GBDT model and the bionic optimization parameter-adjusting model shown in fig. 3, when the real-time electricity price of a future period of time is predicted by using a rolling prediction (prediction is performed every m time periods) method based on the system load rate data obtained by the first step of calculation, first, the real-time electricity price is established according to the system load rate characteristics corresponding to each moment in the period of timeNDepth of plant isDAnd generates corresponding decision trees (i.e., decision tree 1 through decision tree n in the figure), and generates corresponding decision treesNWeak learners (i.e., weak learner 1 through weak learner tree n in the figure) and weight learned by each weak learnerAnd finally, combining all weak learners to form a strong learner algorithm model.
In this embodiment, the system load rate calculated in the previous step is used as a feature training sample, a decision tree is established for the system load rate feature corresponding to each time in the future preset time period to be predicted, and the number of the established decision trees is determinedNCorresponding to the number of system load rate features in a future preset time period. Depth of decision treeDThe training samples can be directly specified according to the characteristics of the training samples, and the method further comprises directly specifying the number of leaf nodes or the depth of the tree, and the number of samples contained in the leaf nodes.
Further, a corresponding weak learner is generated from each decision tree. In this embodiment, the weak learner generated is required to have low variance and high bias, such as CARTTREE (i.e., classification regression tree) may be selected. In the training process, the GBDT model generates a weak classifier through multiple iterations, each iteration generates a weak classifier, each classifier is trained on the basis of the residual error of the last classifier, and the deviation is reduced through iterative training, so that the precision of the final classifier is continuously improved.
It should be noted that, because the present application predicts the real-time electricity price of the power system, the system load rate is selected as the characteristic training sample, and then the weight distribution of the training sample can be initialized, and the training data set with the weight distribution is used for learning, so as to obtain the weak learner. In the training process, when each weak learner learns the weight of the weak learner, as a possible implementation manner, the error rate of each weak learner on the training data set can be calculated, the weight of the current weak learner in the final strong learner is calculated according to the error rate of each weak learner, the larger the error rate is, the smaller the weight is, and the weight learned by each weak learner is further obtained. And finally, carrying out weighted summation on each weak learner to obtain a strong learner, and predicting the real-time electricity price through the strong learner according to the input parameters such as the real-time system load rate and the like.
In an embodiment of the application, in order to improve the generalization capability of the real-time electricity price prediction model, in the process of training the GBDT algorithm model, the bionic optimization parameter-tuning model is further used for optimizing and tuning the number of decision trees, the minimum sample number of the cotyledons of the decision trees, the maximum depth of the decision trees and the maximum characteristic ratio example of data in the GBDT model, so that the real-time electricity price prediction model which is suitable for predicting in different scenes of a power system and has stronger generalization capability is trained.
And S104, correcting the predicted real-time electricity price according to the expected difference value, determining boundary data of energy storage control according to the corrected real-time electricity price, and optimizing a flexible control strategy of the energy storage system based on the boundary data.
Specifically, the real-time electricity price predicted by the model is expected to be corrected according to the difference calculated in step S102, so as to ensure the correctness in the price difference direction, and therefore, the influence of the size and the direction of the price difference between the day-ahead price and the real-time electricity price on the energy storage strategy is comprehensively considered, and the price difference between the day-ahead price and the real-time electricity price is used as the important characteristic data for predicting the real-time electricity price.
Furthermore, boundary data of energy storage control is determined according to the corrected real-time electricity price, and an energy storage flexible control strategy is optimized. Specifically, the flexible control of the energy storage system includes setting that the energy storage device needs to receive a charging and discharging strategy signal at intervals (for example, 15 minutes) to perform specific charging and discharging operations according to a power output curve (for example, from zero point of one day, one point every 15 minutes, and 96 points every day) of an average time interval required by a market, a price curve (for example, from zero point of one day, one point every 15 minutes, and 96 points every day) of an average time interval required by the market, and the like. In order to ensure that the energy storage device can stably operate for a long time and meet the charging and discharging requirements at the current moment, each interval point is required to be rolled and updated with the control strategy, so that the overall optimization is achieved. In the application, the obtained real-time electricity price prediction data is used as the price boundary data of the flexible energy storage control strategy, so that the control parameters such as the energy storage capacity, the charging or discharging operation to be performed, the charging and discharging duration and the like of the energy storage system in the current time period can be determined at each time point (for example, 15 minutes) according to the current predicted real-time electricity price, the energy storage system is controlled according to the determined control parameters in the current time period, and the real-time performance and the accuracy of the optimization result of the flexible control strategy are improved.
In summary, according to the energy storage control method based on rolling type real-time electricity price prediction in the embodiment of the application, the real-time electricity price is predicted in a rolling type prediction mode, the price difference between the current price and the real-time electricity price is used as an important characteristic of the real-time electricity price prediction, and the system load rate is used as input data of the electricity price prediction, so that the prediction accuracy of the real-time electricity price is greatly improved, and a more accurate and reliable price boundary is provided for an energy storage flexible control strategy algorithm. The method logically changes the prediction mode of the prior art at present, better meets the actual operation requirement of an electric field provided with an energy storage device, can provide more accurate data boundary for the optimization of the flexible energy storage control strategy in a rolling mode, and further improves the accuracy of the flexible control strategy. In addition, the method adds a system load rate algorithm model on the design of the algorithm model, establishes a high-efficiency bionic optimization parameter adjusting model to optimize the model parameters, and improves the accuracy and the high efficiency of the real-time electricity price prediction. In addition, the real-time electricity price rolling type prediction algorithm model provided by the method is reasonable and effective in design and high in parameter optimizing speed, and can more fully meet the requirements of the energy storage device in actual operation.
In order to implement the foregoing embodiment, the present application further provides an energy storage control system based on rolling type real-time electricity price prediction, and fig. 4 is a schematic structural diagram of the energy storage control system based on rolling type real-time electricity price prediction according to the embodiment of the present application, and as shown in fig. 4, the system includes a calculation module 100, a statistical analysis module 200, a prediction module 300, and a control module 400.
The calculating module 100 is configured to calculate a rolling time interval of the real-time electricity price prediction.
And the statistical analysis module 200 is configured to obtain actual real-time electricity price data at intervals of the rolling time interval, and perform statistical analysis on the expected difference between the current price and the real-time electricity price through expected statistical analysis.
The prediction module 300 is configured to predict the real-time electricity price in a future preset time period through a rolling prediction algorithm, and includes: and calculating the system load rate at each moment through a preset system load rate algorithm model, and performing real-time electricity price prediction through a gradient lifting decision tree GBDT model based on the system load rate.
And the control module 400 is configured to modify the predicted real-time electricity price according to the expected difference value, determine boundary data of energy storage control according to the modified real-time electricity price, and optimize a flexible control strategy of the energy storage system.
Optionally, in an embodiment of the present application, the computing module 100 is specifically configured to: acquiring a preset energy station declaration output time interval, a spot market clearing price curve time interval and an ultra-short period power prediction value time interval; calculating a rolling time interval for the real-time electricity price prediction by the following formula:
M=min{K,L,N}
wherein M is a rolling time interval of real-time electricity price prediction, K is an energy field station declared output time interval, L is a spot market clearing price curve time interval, and N is an ultra-short period power prediction numerical value time interval.
Optionally, in an embodiment of the present application, the prediction module 300 is specifically configured to calculate the system load rate according to the following formula:
wherein, the first and the second end of the pipe are connected with each other,is composed oftThe rate of system load at the time of day,is composed oftThe predicted value of the load of the whole network at the moment,is composed oftThe predicted value of the new energy output at the moment,is composed oftThe total capacity of all participating spot units at the moment,tindicating any time of day.
Optionally, in an embodiment of the present application, the prediction module 300 is further configured to: establishing decision trees with corresponding number of preset depths according to the system load rate at each moment in a future preset time period; generating a corresponding weak learner according to each decision tree, and acquiring the learned weight of each weak learner; and generating a strong learner algorithm model according to all weak learners and corresponding weights thereof, and predicting the electricity price in real time through the strong learner algorithm model.
Optionally, in an embodiment of the present application, the prediction module 300 is further configured to: and optimizing and adjusting parameters of the number of decision trees, the minimum sample number of cotyledons of the decision trees, the maximum depth of the decision trees and the maximum characteristic number ratio in the GBDT model through a bionic optimization parameter adjusting model.
It should be noted that the foregoing explanation of the embodiment of the energy storage control method based on rolling type real-time electricity price prediction is also applicable to the system of the embodiment, and details are not repeated here
In summary, the energy storage control system based on rolling type real-time electricity price prediction according to the embodiment of the application predicts the real-time electricity price in a rolling type prediction mode, and uses the price difference between the current price and the real-time electricity price as the important characteristic data of the real-time electricity price prediction, so that the prediction accuracy of the real-time electricity price is greatly improved, and a more accurate and reliable price boundary is provided for the energy storage flexible control strategy algorithm. The system better meets the actual operation requirement of an electric field of an energy storage device, can provide more accurate data boundary in a rolling mode for the optimization of the flexible energy storage control strategy, and further improves the accuracy of the flexible control strategy.
In order to achieve the above embodiments, the present application further proposes a non-transitory computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the energy storage control method based on rolling type real-time electricity price prediction as described in any of the above embodiments.
In the description of the present specification, reference to the description of "one embodiment," "some embodiments," "an example," "a specific example," or "some examples" or the like means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
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 to implicitly indicate the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present application.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application, and that variations, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present application.
Claims (6)
1. An energy storage control method based on rolling type real-time electricity price prediction is characterized by comprising the following steps:
calculating a rolling time interval of the real-time electricity price prediction;
acquiring actual real-time electricity price data at intervals of a rolling time interval, and performing statistical analysis on the expected difference between the day-ahead price and the real-time electricity price through expected statistical analysis;
predicting the real-time electricity price in a future preset time period through a rolling type prediction algorithm, wherein the method comprises the following steps: calculating the system load rate of each moment through a preset system load rate algorithm model, and performing real-time electricity price prediction through a gradient lifting decision tree GBDT model based on the system load rate;
correcting the predicted real-time electricity price according to the expected difference value, determining boundary data of energy storage control according to the corrected real-time electricity price, and optimizing a flexible control strategy of the energy storage system based on the boundary data; the calculating of the rolling time interval of the real-time electricity price prediction comprises the following steps:
acquiring a preset energy station declaration output time interval, a spot market clearing price curve time interval and an ultra-short period power prediction value time interval;
calculating a rolling time interval of the real-time electricity rate prediction by the following formula:
M=min{K,L,N}
wherein M is a rolling time interval of real-time electricity price prediction, K is a reporting output time interval of an energy field station, L is a clear price curve time interval of a spot market, and N is an ultra-short period power prediction numerical value time interval; the system load rate algorithm model is shown as the following formula:
wherein the content of the first and second substances,is composed oftThe rate of system load at the time of day,is composed oftThe predicted value of the whole network load at the moment,is composed oftThe predicted value of the new energy output at the moment,is composed oftThe total capacity of all participating spot units at the moment,tindicating any time of day.
2. The control method according to claim 1, wherein the real-time electricity price prediction through a Gradient Boosting Decision Tree (GBDT) model based on the system load rate comprises:
establishing decision trees with corresponding number of preset depths according to the system load rate at each moment in the future preset time period;
generating a corresponding weak learner according to each decision tree, and acquiring the learned weight of each weak learner;
and generating a strong learner algorithm model according to all the weak learners and the corresponding weights thereof, and predicting the electricity price in real time through the strong learner algorithm model.
3. The control method of claim 2, further comprising, prior to said generating a strong learner algorithm model: and optimizing and adjusting parameters of the number of decision trees, the minimum sample number of cotyledons of the decision trees, the maximum depth of the decision trees and the maximum characteristic number ratio in the GBDT model through a bionic optimization parameter adjusting model.
4. An energy storage control system based on rolling type real-time electricity price prediction is characterized by comprising:
the calculation module is used for calculating the rolling time interval of the real-time electricity price prediction;
the statistical analysis module is used for acquiring actual real-time electricity price data at intervals of a rolling time interval and carrying out statistical analysis on the expected difference value between the current price and the real-time electricity price through expected statistical analysis;
the prediction module is used for predicting the real-time electricity price in a future preset time period through a rolling type prediction algorithm and comprises the following steps: calculating the system load rate at each moment through a preset system load rate algorithm model, and predicting the electricity price in real time through a gradient lifting decision tree GBDT model based on the system load rate;
the control module is used for correcting the predicted real-time electricity price according to the expected difference value, determining boundary data of energy storage control according to the corrected real-time electricity price, and optimizing a flexible control strategy of the energy storage system based on the boundary data; the calculation module is specifically configured to:
acquiring a preset energy station declaration output time interval, a spot market clearing price curve time interval and an ultra-short period power prediction value time interval;
calculating a rolling time interval of the real-time electricity rate prediction by the following formula:
M=min{K,L,N}
wherein M is a rolling time interval of real-time electricity price prediction, K is an energy field station declared output time interval, L is a spot market clearing price curve time interval, and N is an ultra-short period power prediction numerical value time interval; the prediction module is specifically configured to calculate a system load rate by the following formula:
wherein the content of the first and second substances,is composed oftThe rate of system load at the time of day,is composed oftThe predicted value of the load of the whole network at the moment,is composed oftThe predicted value of the new energy output at the moment,is composed oftThe total capacity of all participating spot units at the moment,tindicating any time of day.
5. The control system of claim 4, wherein the prediction module is further configured to:
establishing decision trees with corresponding number of preset depths according to the system load rate at each moment in the future preset time period;
generating a corresponding weak learner according to each decision tree, and acquiring the learned weight of each weak learner;
and generating a strong learner algorithm model according to all weak learners and corresponding weights thereof, and predicting the electricity price in real time through the strong learner algorithm model.
6. A non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the energy storage control method according to any one of claims 1-3 based on rolling real-time electricity price prediction.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210880549.9A CN115102202B (en) | 2022-07-25 | 2022-07-25 | Energy storage control method based on rolling type real-time electricity price prediction |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210880549.9A CN115102202B (en) | 2022-07-25 | 2022-07-25 | Energy storage control method based on rolling type real-time electricity price prediction |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115102202A CN115102202A (en) | 2022-09-23 |
CN115102202B true CN115102202B (en) | 2022-11-29 |
Family
ID=83299604
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210880549.9A Active CN115102202B (en) | 2022-07-25 | 2022-07-25 | Energy storage control method based on rolling type real-time electricity price prediction |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115102202B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115659595B (en) * | 2022-09-26 | 2024-02-06 | 中国华能集团清洁能源技术研究院有限公司 | Energy storage control method and device for new energy station based on artificial intelligence |
CN115545800A (en) * | 2022-11-04 | 2022-12-30 | 上海电享信息科技有限公司 | Electric power market trading method and device for virtual power plant and electronic equipment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107563542A (en) * | 2017-08-02 | 2018-01-09 | 阿里巴巴集团控股有限公司 | Data predication method and device and electronic equipment |
EP3640869A1 (en) * | 2018-10-17 | 2020-04-22 | Deutsches Zentrum für Luft- und Raumfahrt e.V. | Method for predicting an energy demand, data processing system and renewable power plant with a storage |
CN113888202A (en) * | 2021-09-03 | 2022-01-04 | 西安峰频能源科技有限公司 | Training method and application method of electricity price prediction model |
-
2022
- 2022-07-25 CN CN202210880549.9A patent/CN115102202B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107563542A (en) * | 2017-08-02 | 2018-01-09 | 阿里巴巴集团控股有限公司 | Data predication method and device and electronic equipment |
EP3640869A1 (en) * | 2018-10-17 | 2020-04-22 | Deutsches Zentrum für Luft- und Raumfahrt e.V. | Method for predicting an energy demand, data processing system and renewable power plant with a storage |
CN113888202A (en) * | 2021-09-03 | 2022-01-04 | 西安峰频能源科技有限公司 | Training method and application method of electricity price prediction model |
Also Published As
Publication number | Publication date |
---|---|
CN115102202A (en) | 2022-09-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN115102202B (en) | Energy storage control method based on rolling type real-time electricity price prediction | |
Shin et al. | Operational planning and optimal sizing of microgrid considering multi-scale wind uncertainty | |
Yang et al. | A deep reinforcement learning method for managing wind farm uncertainties through energy storage system control and external reserve purchasing | |
KR20200100626A (en) | System and method for optimal control of energy storage system | |
Qi et al. | Energyboost: Learning-based control of home batteries | |
CN115496603A (en) | Artificial intelligence technology-based new energy day-ahead transaction decision method for power market | |
CN114696351A (en) | Dynamic optimization method and device for battery energy storage system, electronic equipment and storage medium | |
CN115511634A (en) | New energy day-ahead transaction decision-making method and device for electricity market based on settlement income | |
CN115189416A (en) | Power generation system control method and system based on day-ahead electricity price grading prediction model | |
Zhu et al. | Wind power interval and point prediction model using neural network based multi-objective optimization | |
CN115360741A (en) | Wind storage flexible control method and device based on deep reinforcement learning in spot-cargo scene | |
Rezaeimozafar et al. | A hybrid heuristic-reinforcement learning-based real-time control model for residential behind-the-meter PV-battery systems | |
CN113052630B (en) | Method for configuring electric power equipment by using model and electric power equipment configuration method | |
CN112865235B (en) | Battery control method, electronic device and storage medium | |
CN115619437A (en) | Real-time electricity price determining method and system | |
Avdevicius et al. | Bus charging management based on AI prediction and MILP optimization | |
Ghanbari et al. | Modeling market trading strategies of the intermediary entity for microgrids: A reinforcement learning-based approach | |
CN111754256A (en) | Node electricity price prediction method and device | |
CN115587531B (en) | Segmented solar power limit prediction method and device based on full-network load rate | |
CN115276099B (en) | Flexible control method and device for wind farm energy storage system based on artificial intelligence technology | |
Abdulla et al. | Integrating data-driven forecasting and optimization to improve the operation of distributed energy storage | |
CN115601067A (en) | New energy day-ahead transaction decision-making method and device for electric power market | |
CN115293802A (en) | Electric power transaction auxiliary decision-making method and device for spot market | |
Subramanya et al. | Onsite Renewable Generation Time Shifting for Photovoltaic Systems | |
CN115587890A (en) | Detailed examination loss-based power market new energy day-ahead transaction decision-making 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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |