CN113256016A - Load prediction method and system based on virtual power plant operation - Google Patents
Load prediction method and system based on virtual power plant operation Download PDFInfo
- Publication number
- CN113256016A CN113256016A CN202110625682.5A CN202110625682A CN113256016A CN 113256016 A CN113256016 A CN 113256016A CN 202110625682 A CN202110625682 A CN 202110625682A CN 113256016 A CN113256016 A CN 113256016A
- Authority
- CN
- China
- Prior art keywords
- data
- load prediction
- load
- sub
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000013507 mapping Methods 0.000 claims abstract description 47
- 238000004364 calculation method Methods 0.000 claims abstract description 13
- 238000012549 training Methods 0.000 claims description 33
- 238000012545 processing Methods 0.000 claims description 13
- 230000008569 process Effects 0.000 description 6
- 238000011161 development Methods 0.000 description 3
- 238000013277 forecasting method Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 241001123248 Arma Species 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004836 empirical method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Primary Health Care (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a load prediction method and a load prediction system based on virtual power plant operation, wherein the method comprises the following steps: acquiring related factor data; mapping the relevant factor data into data to be input; inputting data to be input into a pre-trained load prediction model to obtain a load prediction result output by the load prediction model; the load prediction model comprises a sub-load prediction model and a weight value regulation model which are built based on multiple prediction modes; each sub-load prediction model receives data to be input and outputs a sub-prediction result; and the weight value adjusting model receives the sub-prediction results and the data to be input, and performs weighted calculation on the sub-prediction results to obtain load prediction results. According to the scheme, the power system load prediction under different conditions is handled through each sub-load prediction model, the proportion occupied by the prediction result of each sub-load prediction model is adjusted based on the weight value adjusting model, the sub-prediction results are integrated, the load prediction result is obtained, and the prediction precision is improved.
Description
Technical Field
The invention relates to the technical field of load prediction, in particular to a load prediction method and system based on virtual power plant operation.
Background
Accurate load prediction in the power system is one of important means for ensuring safe and stable operation of the system, and meanwhile, important bases are provided for power dispatching, power facility construction and the like of power supply enterprises. There are many factors affecting the power load (e.g., social factors, policy factors, weather factors, holiday and festival factors, etc.). The power load prediction is very complicated, and the prediction result precision is not high.
Disclosure of Invention
In view of this, a load prediction method and system based on virtual power plant operation are provided to solve the problem of low prediction result accuracy in the related art.
The invention adopts the following technical scheme:
the application provides a load prediction method based on virtual power plant operation, includes:
acquiring relevant factor data influencing the load condition in the power system;
mapping the relevant factor data into data to be input based on a preset mapping relation;
inputting the data to be input into a pre-trained load prediction model to obtain a load prediction result output by the load prediction model; the load prediction model comprises a sub-load prediction model and a weight value regulation model which are built based on multiple prediction modes; each sub-load prediction model receives the data to be input and outputs a sub-prediction result; and the weight value adjusting model receives the sub-prediction results and the data to be input, and performs weighted calculation on the sub-prediction results to obtain load prediction results.
Optionally, the method for training the load prediction model includes:
acquiring historical load of a power plant and relevant factor data corresponding to the historical load;
processing the corresponding relevant factor data to obtain corresponding data to be input;
integrating the historical load and the corresponding data to be input as training samples;
training each sub-load prediction model based on the training samples;
and training the weight value adjusting module based on the training sample and the sub-load prediction model completing the training.
Optionally, the relevant factor data includes:
temperature data, rainfall data, wind speed data, relative humidity data, weather type data, week type data, date difference data, and day classification data.
Optionally, the mapping the relevant factor data into data to be input based on a preset mapping relationship includes:
and mapping the data of the different dimension related factors to a specific interval by non-dimensionalization processing.
Optionally, the related factor data includes different categories of data;
the mapping of the non-dimensionalization processing on the data of the related factors of different dimensions to a specific interval comprises the following steps:
and adopting different mapping strategies for the factor data of the same category in different intervals.
Optionally, the method further includes:
acquiring and recording the actual load of the power system;
and retraining the load prediction model at preset time intervals based on the recorded actual load of the power system and corresponding relevant factor data.
Optionally, the method further includes:
and displaying the relevant factor data and the load prediction result based on a preset method.
Optionally, the method further includes: the displaying of the relevant factor data and the load prediction result based on a preset method comprises:
and displaying the load prediction result through a line graph, a bar graph, a pie graph and/or a stacking bar graph.
The application also provides a load prediction system based on virtual power plant operation, includes:
the acquisition module is used for acquiring relevant factor data influencing the load condition in the power system;
the mapping module is used for mapping the relevant factor data into data to be input based on a preset mapping relation;
the prediction module is used for inputting the data to be input into a pre-trained load prediction model to obtain a load prediction result output by the load prediction model; the load prediction model comprises a sub-load prediction model and a weight value regulation model which are built based on multiple prediction modes; each sub-load prediction model receives the data to be input and outputs a sub-prediction result; and the weight value adjusting model receives the sub-prediction results and the data to be input, and performs weighted calculation on the sub-prediction results to obtain load prediction results.
According to the technical scheme, firstly, the data to be input is input into a pre-trained load prediction model to obtain a load prediction result output by the load prediction model; the load prediction model comprises a sub-load prediction model and a weight value regulation model which are built based on multiple prediction modes; each sub-load prediction model receives the data to be input and outputs a sub-prediction result; and the weight value adjusting model receives the sub-prediction results and the data to be input, and performs weighted calculation on the sub-prediction results to obtain load prediction results. Compared with the prior art, in the scheme provided by the application, the load prediction model comprises a sub-load prediction model and a weight value regulation model which are built based on multiple prediction modes. The power system load prediction under different conditions is responded by each sub-load prediction model, the proportion occupied by the prediction result of each sub-load prediction model is adjusted based on the weight adjusting model, so that the sub-prediction results of the sub-load prediction models conforming to the current condition occupy larger weight, the sub-prediction results are integrated to obtain the load prediction result, and the prediction precision is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of a load forecasting method based on virtual power plant operation according to an embodiment of the present invention;
FIG. 2 is a flowchart of a load prediction model in a load prediction method based on virtual power plant operation according to an embodiment of the present invention;
FIG. 3 is a partial flow chart of a load forecasting method based on virtual power plant operation according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a load prediction system based on virtual power plant operation according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail below. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the examples given herein without any inventive step, are within the scope of the present invention.
First, an application scenario of the embodiment of the present invention is explained, where accurate load prediction in a power system is one of important means for ensuring safe and stable operation of the system, and meanwhile, an important basis is provided for power scheduling, power facility construction, and the like of a power supply enterprise. The power load influence factors are numerous (such as social factors, policy factors, meteorological factors, holiday and festival factors, and the like), and the power load prediction is extremely complex. In the prior art, a prediction model is often manufactured through a single prediction mode to predict the power load, but the prediction model manufactured through the single prediction mode cannot adapt to the complexity of a power system, and the problem of low accuracy of a prediction result exists. The present application proposes a corresponding solution to this problem.
Examples
Fig. 1 is a flowchart of a load prediction method based on virtual power plant operation according to an embodiment of the present invention, which may be executed by the load prediction system based on virtual power plant operation according to an embodiment of the present invention. Referring to fig. 1, the method may specifically include the following steps:
s101, acquiring relevant factor data influencing the load condition in the power system;
s102, mapping the relevant factor data into data to be input based on a preset mapping relation;
s103, inputting the data to be input into a pre-trained load prediction model to obtain a load prediction result output by the load prediction model; the load prediction model comprises a sub-load prediction model and a weight value regulation model which are built based on multiple prediction modes; each sub-load prediction model receives the data to be input and outputs a sub-prediction result; and the weight value adjusting model receives the sub-prediction results and the data to be input, and performs weighted calculation on the sub-prediction results to obtain load prediction results.
Compared with the prior art, in the scheme provided by the application, the load prediction model comprises a sub-load prediction model and a weight value regulation model which are built based on multiple prediction modes. The sub-load prediction models correspond to the loads of the power system under different conditions, the proportion occupied by the prediction results of the sub-load prediction models is adjusted based on the weight adjusting model, so that the sub-prediction results of the sub-load prediction models conforming to the current condition occupy larger weight, the sub-prediction results are integrated to obtain the load prediction results, and the prediction precision is improved.
For example, among the plurality of sub-load prediction models, there are a model that is accurate for weekend load prediction and a model that is more accurate for weekday load prediction. When the load of the power system is predicted, the proportion occupied by the prediction result of the sub-load prediction model which can predict the load more accurately in the working day can be increased through the weight adjusting model in the working day; and in weekend, the weight value can be used for adjusting the model, so that the proportion occupied by the prediction result of the sub-load prediction model which can predict the weekend load more accurately is increased.
For a more clear description of the scheme provided in the present application, a specific prediction process is now taken as an example:
firstly, acquiring relevant factor data influencing the load condition in the power system;
specifically, the data of the relevant factors affecting the load condition include, but are not limited to: temperature data, rainfall data, wind speed data, relative humidity data, weather type data, week type data, date difference data, and day classification data. It should be noted that the above data are merely exemplary, and other factors may be included in some specific load prediction scenarios. For example, in some office parks, park occupancy, hours of operation of companies on the park, and the like may be added to the factor-related data.
Further, the related factor data is mapped into data to be input based on a preset mapping relation. In practical applications, because the dimensions of the data (feature quantities) of the relevant factors are different, the values of the different dimensions need to be mapped to a specific interval through non-dimensionalization processing, so that the quantities can have numerical comparability, thereby facilitating the quantitative calculation of the similarity and the difference.
The feature quantities to be considered include two types:
original quantitative index: temperature (maximum temperature, minimum temperature, average temperature, etc.), rainfall, wind speed, relative humidity, etc.
The classification was a quantitative indicator: day weather types (yin, sunny, cloudy, rain, snow, wind, etc.), week types (monday, tuesday,.., sunday, etc.), date differences (days different from the predicted days in historical days, 1 day, 2 days, etc.), day classifications (normal day, birth, festival, etc.), and the like.
When new feature quantities need to be considered, the predictor can join itself. Establishing an index mapping database as the following table:
it should be noted that, since the type of week is a dominant influence factor in the short-term load prediction, the mapping interval can be mapped into the mapping interval of [0.1,3.2] to increase the effect of the week factor, and the mapping values of monday to friday are very close, while saturday and sunday are relatively close, which indicates that monday to friday are normal working days with similar load types, and saturday and sunday are rest days, and there is a relatively large difference between the two groups, which indicates the difference between the working day and the rest day, so that it is advantageous in the clustering analysis. The day classification attribute has a great role in short-term prediction, and particularly has a great influence on a significant festival and days around the significant festival, and if the day classification attribute is not considered, the day classification attribute inevitably has a great influence on a prediction result, so that a great prediction error is generated. In order to treat the normal day and the holiday differently, the prediction of the load curve of the holiday is facilitated.
Furthermore, different mapping strategies are adopted for the factor data of the same category in different intervals. Take the highest temperature as an example. Local high temperature thresholds are assumed to be 30 ° and 35 °. Then for example, a linear mapping may be used between 0 ° and 30 °, with the mapping values varying, but not so much; another group of linear mapping is adopted between 30 degrees and 35 degrees, and the mapping values are obviously different from each other; the non-linear mapping is adopted above 35 degrees, and the mapping value changes greatly when the air temperature increases once. The low temperature region is similar. If the highest daily temperature is the dominant meteorological factor, its mapping interval may exceed the interval limit of [0, 1], while for the non-dominant meteorological factors, its mapping interval should be limited to the [0, 1] interval.
Through the process, after the relevant factor data are mapped into data to be input based on a preset mapping relation, inputting the data to be input into a pre-trained load prediction model to obtain a load prediction result output by the load prediction model; the load prediction model comprises a sub-load prediction model and a weight value regulation model which are built based on multiple prediction modes; each sub-load prediction model receives the data to be input and outputs a sub-prediction result; and the weight value adjusting model receives the sub-prediction results and the data to be input, and performs weighted calculation on the sub-prediction results to obtain load prediction results.
The development rules of the load are different under different conditions, and each prediction method represents a development rule. The more prediction methods, the larger the choice of the predictor, and the more accurate the prediction result. The forecasting personnel can flexibly select a proper forecasting method according to specific conditions, forecasting results of various methods are compared with each other, and then reasonable comprehensive analysis is carried out to obtain a final forecasting result. Methods for short-term load prediction are many, such as multiple regression, spectral analysis, ARMA model, artificial neural network method (ANN), and the like.
In the method, some methods only carry out statistics, analysis and calculation on historical load data, and other related information, particularly weather information with large short-term load influence is not considered, so that the prediction precision of a normal day cannot be further improved, and an error is caused by a special weather day. This is because historical data alone does not reflect the future development trend well, and meteorological factors have a great influence on short-term load and cannot be reflected in the algorithm. Some methods take meteorological factors into consideration, generally adopt empirical methods to compensate by rough meteorological conditions, or enter neural network model calculation as related elements, but the results are sometimes not ideal because the adopted information is too little and the related mode is weak. Furthermore, such methods generally do not involve factors other than weather, and the manner in which the weather factors are accounted for is also inflexible.
Specifically, the prediction method list is as follows:
as shown in fig. 2, in the scheme provided by the present application, the load prediction model includes a sub-load prediction model and a weight adjustment model that are built based on multiple prediction modes; specifically, each of the sub-load prediction models may be constructed according to each of the above principles. By the arrangement, each sub-model can adapt to a complex power load prediction scene. And then, the weight value adjusting model determines the current actual prediction scene based on the data to be input, adjusts the weight occupied by the sub-prediction results of each sub-load prediction model based on the current scene and a preset rule, and integrates each sub-prediction result to obtain a load prediction result.
Further, a training mode of the load prediction model is shown in fig. 3, and specifically includes:
s301, acquiring historical load of a power plant and relevant factor data corresponding to the historical load;
s302, processing the corresponding relevant factor data to obtain corresponding data to be input;
specifically, the data processing mode in step S302 is consistent with the mapping mode of "mapping the relevant factor data to the data to be input based on the preset mapping relationship" in step S102. By the arrangement, the trained model is matched with a scene in actual use, so that the load prediction result is closer to the actual load.
S303, integrating the historical load and the corresponding data to be input as a training sample;
s304, training each sub-load prediction model based on the training samples;
s305, training the weight value adjusting module based on the training sample and the sub-load prediction model completing the training.
The training of the rich deterioration prediction model can be completed through the scheme. It should be noted that, in the scheme provided by the application, only general training steps are described, and specific technical details such as building of a deep learning model, selection of parameters during specific training and the like can be selected according to actual conditions within a certain range and are not repeated one by one again. If the training process is not clear, reference may be made to the prior art training process for the model to facilitate further understanding of the solution provided by the present application.
Further, the scheme provided by the application further comprises:
acquiring and recording the actual load of the power system;
and retraining the load prediction model at preset time intervals based on the recorded actual load of the power system and corresponding relevant factor data.
Therefore, through the retraining mode, the load prediction model cannot be inconsistent with the actual power load prediction scene along with the change of time, and the prediction result can be more reasonable.
And further, displaying the relevant factor data and the load prediction result based on a preset method.
Specifically, the data and results are displayed in a manner that a query request sent by a user is acquired, data is queried based on the query request, the queried data is summarized into a table, and mapping analysis is performed. The generated graph can be switched among four graphs such as a line graph, a bar graph, a pie graph and a stacked bar graph, and can also be switched between two display modes of 2D and 3D. The data in the graph and the table are obtained as seen, and as soon as the data in the data table is changed, the using data is refreshed according to the graph made by the data in the data table, and a new graph is made according to the new data. The drawing can be enlarged and reduced, the color and style can be adjusted, and the drawing can be exported to a local file system of a user as a PNG format picture.
Fig. 4 is a schematic structural diagram of a load prediction system based on virtual power plant operation according to an embodiment of the present invention. Referring to fig. 4, the load prediction system based on virtual power plant operation provided by the present application includes:
an obtaining module 41, configured to obtain data of relevant factors affecting a load condition in an electric power system;
the mapping module 42 is configured to map the relevant factor data into data to be input based on a preset mapping relationship;
the prediction module 43 is configured to input the data to be input into a pre-trained load prediction model to obtain a load prediction result output by the load prediction model; the load prediction model comprises a sub-load prediction model and a weight value regulation model which are built based on multiple prediction modes; each sub-load prediction model receives the data to be input and outputs a sub-prediction result; and the weight value adjusting model receives the sub-prediction results and the data to be input, and performs weighted calculation on the sub-prediction results to obtain load prediction results.
Further, according to the scheme provided by the present application, the method for training the load prediction model includes:
acquiring historical load of a power plant and relevant factor data corresponding to the historical load;
processing the corresponding relevant factor data to obtain corresponding data to be input;
integrating the historical load and the corresponding data to be input as training samples;
training each sub-load prediction model based on the training samples;
and training the weight value adjusting module based on the training sample and the sub-load prediction model completing the training.
Wherein the relevant factor data includes:
temperature data, rainfall data, wind speed data, relative humidity data, weather type data, week type data, date difference data, and day classification data.
The mapping the relevant factor data into data to be input based on a preset mapping relation comprises:
and mapping the data of the different dimension related factors to a specific interval by non-dimensionalization processing.
The correlation factor data comprises different categories of data;
the mapping of the non-dimensionalization processing on the data of the related factors of different dimensions to a specific interval comprises the following steps:
and adopting different mapping strategies for the factor data of the same category in different intervals.
The application provides a load prediction method based on virtual power plant operation still is used for:
acquiring and recording the actual load of the power system;
and retraining the load prediction model at preset time intervals based on the recorded actual load of the power system and corresponding relevant factor data.
And (4) carrying out load prediction results through a line graph, a bar graph, a pie graph and/or a stacking bar graph.
According to the technical scheme, the load prediction system based on virtual power plant operation firstly inputs the data to be input into a pre-trained load prediction model to obtain a load prediction result output by the load prediction model; the load prediction model comprises a sub-load prediction model and a weight value regulation model which are built based on multiple prediction modes; each sub-load prediction model receives the data to be input and outputs a sub-prediction result; and the weight value adjusting model receives the sub-prediction results and the data to be input, and performs weighted calculation on the sub-prediction results to obtain load prediction results. Compared with the prior art, in the scheme provided by the application, the load prediction model comprises a sub-load prediction model and a weight value regulation model which are built based on multiple prediction modes. The sub-load prediction models correspond to the loads of the power system under different conditions, the proportion occupied by the prediction results of the sub-load prediction models is adjusted based on the weight adjusting model, so that the sub-prediction results of the sub-load prediction models conforming to the current condition occupy larger weight, the sub-prediction results are integrated to obtain the load prediction results, and the prediction precision is improved.
It is understood that the same or similar parts in the above embodiments may be mutually referred to, and the same or similar parts in other embodiments may be referred to for the content which is not described in detail in some embodiments.
It should be noted that the terms "first," "second," and the like in the description of the present invention are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. Further, in the description of the present invention, the meaning of "a plurality" means at least two unless otherwise specified.
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 specific logical functions or steps of the process, and alternate implementations are included within the scope of the preferred embodiment of the present invention 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 invention.
It should be understood that portions of the present invention 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. For example, 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 invention 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.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean 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 invention. In this specification, the schematic representations of the terms used above do not necessarily 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.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (9)
1. A load prediction method based on virtual power plant operation is characterized by comprising the following steps:
acquiring relevant factor data influencing the load condition in the power system;
mapping the relevant factor data into data to be input based on a preset mapping relation;
inputting the data to be input into a pre-trained load prediction model to obtain a load prediction result output by the load prediction model; the load prediction model comprises a sub-load prediction model and a weight value regulation model which are built based on multiple prediction modes; each sub-load prediction model receives the data to be input and outputs a sub-prediction result; and the weight value adjusting model receives the sub-prediction results and the data to be input, and performs weighted calculation on the sub-prediction results to obtain load prediction results.
2. The virtual power plant operation based load prediction method of claim 1, wherein the method of training the load prediction model comprises:
acquiring historical load of a power plant and relevant factor data corresponding to the historical load;
processing the corresponding relevant factor data to obtain corresponding data to be input;
integrating the historical load and the corresponding data to be input as training samples;
training each sub-load prediction model based on the training samples;
and training the weight value adjusting module based on the training sample and the sub-load prediction model completing the training.
3. The virtual power plant operation based load prediction method of claim 1, wherein the relevant factor data comprises:
temperature data, rainfall data, wind speed data, relative humidity data, weather type data, week type data, date difference data, and day classification data.
4. The virtual power plant operation based load prediction method according to claim 1, wherein the mapping the relevant factor data to be input based on a preset mapping relation comprises:
and carrying out non-dimensionalization processing on the data of the different dimension related factors to map to a specific interval.
5. The virtual power plant operation based load prediction method of claim 4, wherein the relevant factor data comprises different categories of data;
the non-dimensionalization processing of the data of the related factors of different dimensions to map to a specific interval comprises the following steps:
and adopting different mapping strategies for the factor data of the same category in different intervals.
6. The virtual power plant operation based load prediction method of claim 1, further comprising:
acquiring and recording the actual load of the power system;
and retraining the load prediction model at preset time intervals based on the recorded actual load of the power system and corresponding relevant factor data.
7. The virtual power plant operation based load prediction method of claim 1, further comprising:
and displaying the relevant factor data and the load prediction result based on a preset method.
8. The virtual power plant operation based load prediction method of claim 7, further comprising: the displaying of the relevant factor data and the load prediction result based on a preset method comprises:
and displaying the load prediction result through a line graph, a bar graph, a pie graph and/or a stacking bar graph.
9. A load prediction system based on virtual power plant operation, comprising:
the acquisition module is used for acquiring relevant factor data influencing the load condition in the power system;
the mapping module is used for mapping the relevant factor data into data to be input based on a preset mapping relation;
the prediction module is used for inputting the data to be input into a pre-trained load prediction model to obtain a load prediction result output by the load prediction model; the load prediction model comprises a sub-load prediction model and a weight value regulation model which are built based on multiple prediction modes; each sub-load prediction model receives the data to be input and outputs a sub-prediction result; and the weight value adjusting model receives the sub-prediction results and the data to be input, and performs weighted calculation on the sub-prediction results to obtain load prediction results.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110625682.5A CN113256016A (en) | 2021-06-04 | 2021-06-04 | Load prediction method and system based on virtual power plant operation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110625682.5A CN113256016A (en) | 2021-06-04 | 2021-06-04 | Load prediction method and system based on virtual power plant operation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113256016A true CN113256016A (en) | 2021-08-13 |
Family
ID=77186410
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110625682.5A Pending CN113256016A (en) | 2021-06-04 | 2021-06-04 | Load prediction method and system based on virtual power plant operation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113256016A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108416366A (en) * | 2018-02-06 | 2018-08-17 | 武汉大学 | A kind of power-system short-term load forecasting method of the weighting LS-SVM based on Meteorological Index |
| CN108898246A (en) * | 2018-06-19 | 2018-11-27 | 深圳供电局有限公司 | A kind of load prediction management system based on electric system |
| CN108985570A (en) * | 2018-08-17 | 2018-12-11 | 深圳供电局有限公司 | A kind of load forecasting method and its system |
| CN110675275A (en) * | 2019-09-05 | 2020-01-10 | 深圳供电局有限公司 | Demand side response power load regulation and control method and system of virtual power plant |
-
2021
- 2021-06-04 CN CN202110625682.5A patent/CN113256016A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108416366A (en) * | 2018-02-06 | 2018-08-17 | 武汉大学 | A kind of power-system short-term load forecasting method of the weighting LS-SVM based on Meteorological Index |
| CN108898246A (en) * | 2018-06-19 | 2018-11-27 | 深圳供电局有限公司 | A kind of load prediction management system based on electric system |
| CN108985570A (en) * | 2018-08-17 | 2018-12-11 | 深圳供电局有限公司 | A kind of load forecasting method and its system |
| CN110675275A (en) * | 2019-09-05 | 2020-01-10 | 深圳供电局有限公司 | Demand side response power load regulation and control method and system of virtual power plant |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10860615B2 (en) | Users campaign for peaking energy usage | |
| CN111932402B (en) | Short-term power load bidirectional combination prediction method based on similar day and LSTM | |
| CN109657844B (en) | Electric power short-term load prediction method and device | |
| Goude et al. | Local short and middle term electricity load forecasting with semi-parametric additive models | |
| Khotanzad et al. | A neuro-fuzzy approach to short-term load forecasting in a price-sensitive environment | |
| CN110503256B (en) | Short-term load prediction method and system based on big data technology | |
| CN113962364A (en) | Multi-factor power load prediction method based on deep learning | |
| CN111680841B (en) | Short-term load prediction method, system and terminal equipment based on principal component analysis | |
| CN112561207A (en) | Power grid load prediction method and device considering meteorological factors and power system | |
| CN112330077B (en) | Power load prediction method, power load prediction device, computer equipment and storage medium | |
| CN110675275A (en) | Demand side response power load regulation and control method and system of virtual power plant | |
| CN115470862A (en) | Dynamic self-adaptive load prediction model combination method | |
| Qiao et al. | Predicting building energy consumption based on meteorological data | |
| CN113902181A (en) | Short-term prediction method and equipment for common variable heavy overload | |
| CN113256472A (en) | Intelligent traffic control method and system and brain-like computer readable storage medium | |
| JP2000270473A (en) | Power demand estimating method | |
| CN113570106A (en) | Power load prediction method and device based on expressway neural network | |
| CN117374920A (en) | Ultra-short-term prediction method, device and medium considering environmental factors | |
| CN113256016A (en) | Load prediction method and system based on virtual power plant operation | |
| CN116934366A (en) | Charging pricing method and device for charging station, storage medium and computer equipment | |
| CN116826710A (en) | Peak clipping strategy recommendation method and device based on load prediction and storage medium | |
| Souzanchi-K et al. | A multi adaptive neuro fuzzy inference system for short term load forecasting by using previous day features | |
| CN115115107A (en) | Photovoltaic power prediction method and device and computer equipment | |
| CN112348235A (en) | Wind-solar bus load self-adaptive prediction method and device and computer equipment | |
| Farhadi et al. | A fuzzy inference self-organizing-map based model for short term load forecasting |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210813 |
|
| RJ01 | Rejection of invention patent application after publication |