Regional power prediction method and device, storage medium and electronic device

By performing multi-dimensional evaluation of the actual measurement and predicted power data of the station, stations with significant impact on regional power and low similarity were selected, and multiple prediction models were established for combined prediction, which solved the single and common screening indicators in the existing technology. The problem of limited nature improves the accuracy and stability of regional power prediction.

CN119093348APending Publication Date: 2024-12-06HUANENG CLEAN ENERGY RES INST
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411194157.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-28
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

In the prior art, there is a problem of how to screen suitable for stations with appropriate numbers and characteristics to establish regional lift-scale power prediction models, and there is a problem of single screening indicators and limited universality.

Method used

By obtaining the measured power data, predicted power data and regional measured power data of each station, calculating the evaluation score, prioritizing the scores, filtering out the stations with significant impact on regional power and low similarity, and establishing multiple predictions Model and combine predictions to achieve accurate prediction of regional power.

Benefits of technology

Effectively screening out stations with significant impact on regional power and low similarity improves the accuracy and stability of regional power prediction, and overcomes the problems of single screening indicators and limited versatility in the prior art.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119093348A_ABST
    Figure CN119093348A_ABST
Patent Text Reader

Abstract

The invention discloses a regional power prediction method and device, a storage medium and an electronic device, and the method comprises the steps: obtaining the power data of each station in a target region, the power data at least comprises actually measured power data of each station, predicted power data of each station and regional actually measured power data of a region where each station is located; determining an evaluation score of the power data on the area power data of the target area; according to the value of the evaluation score, all stations in the target area are subjected to priority ranking, a group of stations in all the stations are determined according to the ranking result, the group of stations are multiple target stations existing in the target area, and the target value of the influence of the multiple target stations on the area power is larger than a preset value; the similarity between the plurality of target stations and other stations is smaller than a preset similarity; and performing upscaling power prediction on the target area based on the group of stations.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of electric power, and in particular, to a method and device for predicting regional power, a storage medium and an electronic device. Background Art

[0002] In the new power system, the proportion of new energy power generation represented by wind power and photovoltaic power generation is increasing. The randomness and volatility of new energy output have put forward new requirements for grid dispatching and stable operation of power stations. Accurate, efficient and reliable power prediction methods are the key to solving this problem. According to the different scales of the prediction object, the power prediction method can be divided into unit level, station level, group level, regional level, etc. At present, the station-level power prediction algorithm is mainly developed and promoted, but with the increase in the number of new energy stations, the importance of regional power prediction technology has gradually become prominent. At present, there are two main regional power prediction methods: one is to add the power prediction data of all wind farms and photovoltaic power stations in the region to obtain the overall prediction result, but the quality of station-level data is very high; the other is to establish a model from a few stations to regional power prediction by screening stations with specific indicators, but there are problems such as single screening indicators and limited versatility.

[0003] Regarding the problem of how to select appropriate number and characteristics of stations to establish a regional upscaling power prediction model in related technologies, no effective solution has been proposed so far.

[0004] Therefore, it is necessary to improve the related technology to overcome the above-mentioned defects in the related technology. Summary of the invention

[0005] The embodiments of the present application provide a method and device for predicting regional power, a storage medium, and an electronic device to at least solve the problem of how to select sites with appropriate numbers and characteristics to establish a regional upscaled power prediction model.

[0006] According to one aspect of an embodiment of the present application, a method for predicting regional power is provided, comprising: obtaining power data of each station in a target area, wherein the power data comprises at least: measured power data of each station, predicted power data of each station, and regional measured power data of the area where each station is located; determining an evaluation score of the power data to regional power data of the target area; prioritizing all stations in the target area according to the numerical values ​​of the evaluation scores, and determining a group of stations from all stations through the sorting results, wherein a group of stations is a plurality of target stations existing in the target area, the target values ​​of the plurality of target stations affecting the regional power are greater than a preset value, and the similarity between the plurality of target stations and other stations is less than a preset similarity; and predicting the upscaled power of the target area based on the group of stations.

[0007] In an exemplary embodiment, an upscaled power prediction of a target area is performed based on a group of stations, including: determining a plurality of target measured power data corresponding to the plurality of target stations and regional actual power data corresponding to the plurality of target stations; determining the plurality of target measured power data as model inputs and the regional actual power data as model outputs to establish a training set; wherein a corresponding relationship exists between the target measured power data and the regional actual power data; using the training set to train a target prediction model selected from a target model database to obtain a plurality of prediction models; and predicting the upscaled power of the target area through the plurality of prediction models.

[0008] In an exemplary embodiment, upscaled power is predicted for a target area using multiple prediction models, including: identifying a site type corresponding to each prediction model in the multiple prediction models; determining a prediction weight corresponding to each prediction model according to the site type; combining the multiple prediction models based on the prediction weights to obtain a combined prediction model for upscaling power for the target area; and predicting the upscaled power for the target area using the combined prediction model.

[0009] In an exemplary embodiment, determining an evaluation score of power data for regional power data of a target area includes: obtaining a first time series of measured power data of each station, and a second time series of regional measured power data of the area where each station is located; determining a second correlation score for indicating accuracy of each station based on the measured power data and predicted power data corresponding to the measured power data; when the installed capacity of each station is obtained, determining a third correlation score for indicating the proportion of installed capacity of each station based on the installed capacity of each station and the total regional installed capacity corresponding to the target area; and determining an evaluation score based on the first correlation score, the second correlation score, and the third correlation score.

[0010] In an exemplary embodiment, an evaluation score is determined according to a first relevance score, a second relevance score, and a third relevance score, including: when determining the first relevance score, the second relevance score, and the third relevance score corresponding to different stations, sorting the different stations on different relevance dimensions according to the size of the scores to obtain ranking values ​​corresponding to each relevance dimension for the different stations; when determining the first ranking value, the second ranking value, and the third ranking value corresponding to each station, taking a weighted average of the first ranking value, the second ranking value, and the third ranking value to obtain an evaluation score corresponding to each station, wherein the weights corresponding to the weighted average are respectively a first weight corresponding to the first ranking value, a second weight corresponding to the second ranking value, and a third weight corresponding to the third ranking value.

[0011] In an exemplary embodiment, all stations in the target area are prioritized according to the numerical values ​​of the evaluation scores, and a group of stations among all the stations is determined through the sorting results, including: obtaining the sorting results of the priority sorting; selecting the first station with the highest priority in the sorting results to add to a preset station set; calculating the geographical location similarity between the first station and the station with the highest priority among the remaining stations in the sorting results; determining other stations to be added to the station set based on the relationship between the geographical location similarity and a preset location threshold; and determining a group of stations among all the stations based on the station set when it is determined that the location judgment of all stations has been completed.

[0012] According to another aspect of an embodiment of the present application, a regional power prediction device is also provided, including: an acquisition module, used to acquire power data of each station in a target area, wherein the power data includes at least: measured power data of each station, predicted power data of each station, and regional measured power data of the area where each station is located; a determination module, used to determine an evaluation score of the power data to the regional power data of the target area; a sorting module, used to prioritize all stations in the target area according to the numerical values ​​of the evaluation scores, and determine a group of stations among all stations through the sorting results, wherein a group of stations is a plurality of target stations existing in the target area, the target values ​​of the impact of the plurality of target stations on the regional power are greater than a preset value, and the similarity between the plurality of target stations and other stations is less than a preset similarity; a prediction module, used to predict the upscaled power of the target area based on a group of stations.

[0013] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned regional power prediction method when running.

[0014] According to another aspect of an embodiment of the present application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the regional power prediction method through the computer program.

[0015] According to another aspect of the embodiments of the present application, a computer program product is provided, including a computer program, and the method for predicting regional power when the computer program is executed by a processor.

[0016] Through this application, the power data of each station in the target area is obtained, wherein the power data at least includes: the measured power data of each station, the predicted power data of each station, and the regional measured power data of the area where each station is located; determine the evaluation score of the power data to the regional power data of the target area; prioritize all stations in the target area according to the numerical value of the evaluation score, and determine a group of stations among all stations through the sorting result, wherein a group of stations is a plurality of target stations existing in the target area, the target values ​​of the impact of the plurality of target stations on the regional power are greater than the preset value, and the similarity between the plurality of target stations and other stations is less than the preset similarity; predict the upscaled power of the target area based on a group of stations. The above technical solution is adopted to solve the problem of how to select stations with appropriate number and characteristics to establish a regional upscaled power prediction model. Furthermore, by scoring and sorting the power data of different stations, a group of stations with significant impact on the power of the target area and low similarity are determined for predicting the regional power. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0019] Figure 1 It is a hardware structure block diagram of a computer terminal of a method for predicting regional power in an embodiment of the present application;

[0020] Figure 2 is a flow chart of a method for predicting regional power according to an embodiment of the present application;

[0021] Figure 3 It is a flow chart of a regional power prediction upscaling method based on multi-dimensional evaluation of new energy stations according to an embodiment of the present application;

[0022] Figure 4 A flow chart of a site screening process according to an embodiment of the present application;

[0023] Figure 5 It is a structural block diagram of regional power prediction according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] The method embodiments provided in the embodiments of the present application can be executed in a computer terminal, a mobile terminal or a similar computing device. Taking running on a computer terminal as an example, Figure 1 FIG. 1 is a hardware structure block diagram of a computer terminal of a method for predicting regional power in an embodiment of the present application. Figure 1 As shown, the computer terminal may include one or more ( Figure 1 The computer terminal includes a processor 102 (only one is shown in the figure) (the processor 102 may include but is not limited to a processing device such as a microprocessor (Central Processing Unit, CPU) or a programmable logic device (Field Programmable Gate Array, FPGA)) and a memory 104 for storing data, wherein the computer terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above-mentioned computer terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0027] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the regional power prediction method in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0028] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of a computer terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0029] In this embodiment, a method for predicting regional power is provided. Figure 2 is a flow chart of a method for predicting regional power according to an embodiment of the present application, such as Figure 2 As shown, the process includes the following steps S202-S208:

[0030] Step S202: acquiring power data of each station in the target area, wherein the power data at least includes: measured power data of each station, predicted power data of each station, and regional measured power data of the area where each station is located;

[0031] Step S204: determining an evaluation score of the power data with respect to the regional power data of the target area;

[0032] Step S206: Prioritizing all stations in the target area according to the numerical values ​​of the evaluation scores, and determining a group of stations from all the stations through the ranking results, wherein the group of stations is a plurality of target stations existing in the target area, the target numerical values ​​of the plurality of target stations on the regional power are greater than a preset numerical value, and the similarity between the plurality of target stations and other stations is less than a preset similarity;

[0033] Step S208: predicting the upscaled power of the target area based on the group of sites.

[0034] The above steps obtain the power data of each station in the target area, wherein the power data at least includes: the measured power data of each station, the predicted power data of each station, and the regional measured power data of the area where each station is located; determine the evaluation score of the power data to the regional power data of the target area; prioritize all stations in the target area according to the numerical value of the evaluation score, and determine a group of stations among all stations through the sorting results, wherein a group of stations is a plurality of target stations existing in the target area, the target values ​​of the impact of the plurality of target stations on the regional power are greater than the preset value, and the similarity between the plurality of target stations and other stations is less than the preset similarity; predict the upscaled power of the target area based on a group of stations. The above technical solution solves the problem of how to select stations with appropriate number and characteristics to establish a regional upscaled power prediction model. Furthermore, by scoring and sorting the power data of different stations, a group of stations with significant impact on the power of the target area and low similarity are determined for predicting the regional power.

[0035] In an exemplary embodiment, an upscaled power prediction of a target area is performed based on a group of stations, including: determining a plurality of target measured power data corresponding to the plurality of target stations and regional actual power data corresponding to the plurality of target stations; determining the plurality of target measured power data as model inputs and the regional actual power data as model outputs to establish a training set; wherein a corresponding relationship exists between the target measured power data and the regional actual power data; using the training set to train a target prediction model selected from a target model database to obtain a plurality of prediction models; and predicting the upscaled power of the target area through the plurality of prediction models.

[0036] In an exemplary embodiment, the upscaled power of the target area is predicted using the multiple prediction models, including: identifying the station type corresponding to each prediction model in the multiple prediction models; determining the prediction weight corresponding to each prediction model according to the station type; combining the multiple prediction models based on the prediction weights to obtain a combined prediction model for upscaling the power of the target area; and predicting the upscaled power of the target area using the combined prediction model.

[0037] Optionally, for a target area, upscaled power prediction is performed, and three different types of prediction models are selected: linear regression model, support vector machine model, and neural network model. First, the station type corresponding to each prediction model is identified, such as the linear regression model is suitable for stations with simple linear relationships, the support vector machine model is suitable for stations with nonlinear relationships, and the neural network model is suitable for complex stations. Then, the prediction weight of each prediction model is determined according to the station type, such as the linear regression model weight is 0.3, the support vector machine model weight is 0.5, and the neural network model weight is 0.2. Next, the three prediction models are combined based on the prediction weights to obtain a combined prediction model for upscaling power in the target area. Finally, the power of the target area is predicted by combining the prediction models to obtain more accurate and reliable prediction results, providing better reference and decision support for practical applications. In this way, the advantages of different models can be fully utilized to improve the accuracy and stability of prediction.

[0038] In an exemplary embodiment, determining the evaluation score of the power data for the regional power data of the target area includes: obtaining a first time series of the measured power data of each station and a second time series of the regional measured power data of the area where each station is located; calculating a first correlation score between the first time series and the second time series in time series; determining a second correlation score for indicating accuracy of each station based on the measured power data and the predicted power data corresponding to the measured power data; when the installed capacity of each station is obtained, determining a third correlation score for indicating the proportion of installed capacity of each station based on the installed capacity of each station and the total regional installed capacity corresponding to the target area; and determining the evaluation score based on the first correlation score, the second correlation score, and the third correlation score.

[0039] Optionally, the calculation of the first correlation score between the first time series and the second time series in the above embodiment refers to calculating the Pearson correlation coefficient R between the actual power of each station and the actual power time series of the region. i .

[0040] Exemplarily, the target area includes three stations A, B, and C, and the installed capacity of each station is 100MW, 150MW, and 200MW, respectively. The measured power data of each station and the regional measured power data of the area where it is located are obtained, and the relevant scores are calculated as follows: the correlation score between station A and the target area is 0.8, station B is 0.7, and station C is 0.6. The accuracy correlation score is calculated based on the measured power data of each station and the predicted power data corresponding to each station: station A is 0.9, station B is 0.8, and station C is 0.7. The relevant score is calculated based on the proportion of installed capacity: station A is 0.3, station B is 0.4, and station C is 0.5. Finally, the evaluation score is calculated based on the above three relevant scores. In this way, the performance of each station in the regional power data can be objectively evaluated. The higher the correlation coefficient between the station and the regional power, the higher the accuracy of the station power prediction, the greater the proportion of the station installed capacity, and the greater the influence of the power prediction result of the station on the power prediction result of the region.

[0041] In an exemplary embodiment, the evaluation score is determined according to the first relevance score, the second relevance score, and the third relevance score, including: when determining the first relevance score, the second relevance score, and the third relevance score corresponding to different stations, sorting the different stations in different relevance dimensions according to the size of the scores to obtain the ranking values ​​corresponding to each relevance dimension of the different stations; when determining the first ranking value, the second ranking value, and the third ranking value corresponding to each station, weighted averaging the first ranking value, the second ranking value, and the third ranking value to obtain the evaluation score corresponding to each station, wherein the weights corresponding to the weighted average are respectively the first weight corresponding to the first ranking value, the second weight corresponding to the second ranking value, and the third weight corresponding to the third ranking value.

[0042] Optionally, there are three sites A, B, and C, and their rankings on the first relevance score, the second relevance score, and the third relevance score are as follows:

[0043] Station A: The first relevant score ranks 1, the second relevant score ranks 3, and the third relevant score ranks 2;

[0044] Station B: The first relevant score ranked 3, the second relevant score ranked 2, and the third relevant score ranked 1;

[0045] Station C: The first relevant score ranked 2, the second relevant score ranked 1, and the third relevant score ranked 3;

[0046] Set the weights to 0.4 for the first ranking value, 0.3 for the second ranking value, and 0.3 for the third ranking value, and calculate the evaluation score:

[0047] Station A: (1*0.4+3*0.3+2*0.3)=1.9;

[0048] Station B: (3*0.4+2*0.3+1*0.3)=2.1;

[0049] Station C: (2*0.4+1*0.3+3*0.3)=2.0;

[0050] According to the calculation results, it can be seen that site A performed best in the overall evaluation, followed by site C, and site B ranked lowest. This evaluation method can help managers understand the performance of each site more intuitively and formulate improvement plans in a targeted manner.

[0051] In an exemplary embodiment, all stations in the target area are prioritized according to the numerical values ​​of the evaluation scores, and a group of stations among all the stations is determined through the ranking results, including: obtaining the ranking results of the priority ranking; selecting the first station with the highest priority in the ranking results to add to a preset station set; calculating the geographical location similarity between the first station and the station with the highest priority among the remaining stations in the ranking results; determining other stations to be added to the station set based on the relationship between the geographical location similarity and a preset location threshold; and when it is determined that the position judgment of all stations has been completed, determining a group of stations among all the stations based on the station set.

[0052] Optionally, according to the above embodiment, first obtain the priority sorting result, station A, station B, station C, station D; then, select station A with the highest priority to join the preset station set; then, calculate the geographical location similarity between station A and the remaining stations B, C, and D, and obtain similarities of 0.8, 0.6, and 0.4 respectively; further, set the preset location threshold to 0.5, and according to the relationship between the similarity and the threshold, add the stations less than the preset threshold to the set, and obtain the result of adding station D to the station set; finally, after completing the position judgment of all stations, determine a set of stations A and D. The above technical solution can obtain the most suitable and optimal set of stations by prioritizing the stations and calculating the geographical location similarity, thereby improving the efficiency and practicality of the solution.

[0053] Obviously, the embodiments described above are only some embodiments of the present application, not all embodiments. In order to better understand the above method, the above process is described below in conjunction with the embodiments, but it is not intended to limit the technical solutions of the embodiments of the present application, specifically:

[0054] The optional embodiment of the present application provides a regional power prediction upscaling method based on multi-dimensional evaluation of new energy stations to solve the problem of how to screen stations with appropriate numbers and characteristics to establish a regional upscaling power prediction model. The method comprehensively evaluates the importance of the power data of a single station to the regional power data in multiple dimensions through indicators such as correlation coefficient, accuracy, and installed capacity, and proposes a method for generating a station priority queue, so as to more effectively select stations with higher power data quality and importance for power prediction; at the same time, the similarity index is used to evaluate the degree of information duplication between the station and the existing station set, and a method for screening the station set is proposed to further improve the prediction accuracy. Finally, based on the multi-dimensional evaluation and similarity screening method, an upscaling power prediction method is established to achieve accurate prediction of regional power data through a small number of station data.

[0055] Optional, Figure 3 It is a flow chart of a regional power prediction upscaling method based on multi-dimensional evaluation of new energy stations according to an embodiment of the present application; specifically comprising the following steps:

[0056] Step 1: Data acquisition. In the data acquisition part, for stations with complete data in the region, it is necessary to obtain measured power data and predicted power data for calculating accuracy, and regional-level measured power data is also required. All power time series need to be aligned in time.

[0057] Step 2: Sorting of evaluation indicators. In the evaluation indicator ranking part, it is necessary to calculate the multi-dimensional evaluation indicators of each station and evaluate the influence of each station on the regional power data. The specific steps include step 2.1-step 2.4:

[0058] Step 2.1: Calculate the Pearson correlation coefficient R between the actual power of each station and the regional actual power time series i .

[0059]

[0060] Where P i (t) is the measured power data of the ith station at time t, P(t) is the measured power data of the region at time t, n is the length of the time series, Represents the arithmetic mean.

[0061] Step 2.2: Calculate the evaluation index of power prediction of each station according to the grid connection rules of the province where the region is located, usually the accuracy or qualification rate index. Taking the common accuracy calculation method as an example, the calculation formula is as follows:

[0062]

[0063] Among them A i is the power prediction accuracy of the i-th station, Pi,p (t) is the predicted power data of the ith station at time t, cap i is the installed capacity of the ith station, and n is the number of stations.

[0064] Step 2.3: Calculate the installed capacity share of each station α i .

[0066]

[0067] Where cap is the total installed capacity of the region.

[0068] Step 2.4: Set the weights of the evaluation indicators and establish the priority ranking of each station. The evaluation indicators such as correlation coefficient, accuracy, and installed capacity share all reflect the importance of the correlation between a single station and regional power. That is, the higher the correlation coefficient between the station and regional power, the higher the accuracy of the station power prediction, the larger the proportion of the station installed capacity, and the greater the impact of the station's power prediction results on the regional power prediction results.

[0069] First, the R i , A i and α i After sorting them separately, three groups of ranking sequences are obtained. Then the rankings of the three indicators of each station are weighted averaged to obtain the multi-dimensional indicator evaluation score of the station. The weight coefficient of each indicator can be set according to the difference between the stations. If the difference of a certain indicator between the stations is small, a low weight should be set, otherwise a high weight should be set. Finally, the multi-dimensional indicator evaluation scores of each station are arranged in descending order to obtain the priority queue of stations participating in regional upscaling prediction.

[0070] Step 3: Site selection. In the site selection part, it is necessary to select several sites with low similarity from the obtained site priority queue according to the similarity between sites to form a site set for regional upscaling power prediction. Figure 4 As shown, Figure 4 It is a flow chart of the site screening process according to an embodiment of the present application; the specific steps include steps 3.1 to 3.4:

[0071] Step 3.1: Add the first station in the station priority queue to the station set.

[0072] Step 3.2: For the station with the highest priority among the remaining stations, calculate the geographical location similarity between the station and the stations in the existing station set. The geographical location similarity between two stations is calculated as follows:

[0073]

[0074] Where lon is the longitude of the station, lat is the latitude of the station, and the subscripts i and j represent two different stations. Calculate the geographical location similarity between the station and each station in the existing station set, and find the average geographical location similarity. If the average similarity is higher than the set threshold δ a , then the station is not added to the station set, and step 3.2 is repeated; if the average similarity is lower than the set threshold δ a , then go to step 3.3.

[0075] Step 3.3: For the station in step 2, calculate the power time series similarity between the station and the stations in the existing station set. The power time series similarity between two stations is calculated as follows:

[0076]

[0077] Calculate the power time series similarity between the station and each station in the existing station set, and find the average value of the time series similarity. If the average value of the similarity is higher than the set threshold δ b , then the station is not added to the station set and returns to step 3.2; if the average similarity is lower than the set threshold δ b , then the station meets the requirement of low similarity with the station set, and the station is added to the station set.

[0078] Step 3.4: Repeat steps 3.2 and 3.3 until all sites have gone through the above screening process.

[0079] Step 4: Regional upscaling prediction. The selected station set includes a group of stations with great impact on regional power and low similarity between stations. The station in the station set can be used to perform regional upscaling power prediction. The main steps include step 4.1-step 4.3:

[0080] Step 4.1: Take the measured power data of each station in the station set as input and the actual power data of the region as output to establish a regional upscaling power prediction model. Considering the relationship between the station and the regional power, after the above screening process, a good upscaling prediction accuracy can be obtained without a complex model. Models that can be used for training include linear regression models, support vector regression models, multi-layer perceptron models, etc. A more complex machine learning model can be selected according to the model prediction effect.

[0081] Step 4.2: For the multiple prediction models trained in step 4.1, set the weights of individual models to obtain a combined prediction model.

[0082] Step 4.3: Using the prediction data of each station in the station set as input, the combined prediction model obtained in step 4.2 can be used to calculate the final regional upscaled power prediction result.

[0083] It should be noted that the regional upscaling power prediction requires selecting stations with a high degree of correlation with the regional output from multiple stations in the region. At the same time, in order to avoid information redundancy and increase the amount of calculation, the information duplication contained in multiple stations should be as low as possible, while ensuring the accuracy and speed of the prediction. If the information duplication contained in multiple stations is high, then when performing regional upscaling power prediction, redundant information will be repeatedly calculated, thereby increasing the amount of calculation and calculation time. In addition, if the information redundancy is high, the prediction result may be interfered by the redundant information, resulting in a decrease in the accuracy of the prediction. Therefore, it is necessary to pay attention to information redundancy to ensure the accuracy of the prediction.

[0084] In summary, the optional embodiments of the present application comprehensively evaluate the importance of the power data of a single station to the regional power data through multiple dimensions such as correlation coefficient, accuracy, installed capacity and other indicators, and propose a method for generating a station priority queue, so as to more effectively select stations with higher power data quality and importance for power prediction; at the same time, the similarity index is used to evaluate the degree of information duplication between the station and the existing station set, and a method for screening the station set is proposed to further improve the prediction accuracy; finally, based on the multi-dimensional evaluation and similarity screening method, an upscaling power prediction method is established to achieve accurate prediction of regional-level power data through data from a few stations.

[0085] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0086] In this embodiment, a regional power prediction device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0087] Figure 5 : is a structural block diagram of a regional power prediction device according to an embodiment of the present application, the device comprising:

[0088] An acquisition module 52 is used to acquire power data of each station in the target area, wherein the power data at least includes: measured power data of each station, predicted power data of each station, and regional measured power data of the area where each station is located;

[0089] A determination module 54, configured to determine an evaluation score of the power data with respect to the regional power data of the target area;

[0090] A sorting module 56 is used to prioritize all the stations in the target area according to the numerical values ​​of the evaluation scores, and determine a group of stations from all the stations through the sorting results, wherein the group of stations is a plurality of target stations existing in the target area, the target numerical values ​​of the plurality of target stations on the regional power are greater than a preset numerical value, and the similarity between the plurality of target stations and other stations is less than a preset similarity;

[0091] The prediction module 58 is used to predict the upscaled power of the target area based on the group of sites.

[0092] The above device obtains the power data of each station in the target area, wherein the power data at least includes: the measured power data of each station, the predicted power data of each station, and the regional measured power data of the area where each station is located; determines the evaluation score of the power data to the regional power data of the target area; prioritizes all stations in the target area according to the numerical value of the evaluation score, and determines a group of stations among all stations through the sorting result, wherein a group of stations is a plurality of target stations existing in the target area, the target values ​​of the impact of the plurality of target stations on the regional power are greater than the preset value, and the similarity between the plurality of target stations and other stations is less than the preset similarity; and predicts the upscaled power of the target area based on a group of stations. The above technical solution solves the problem of how to select stations with appropriate number and characteristics to establish a regional upscaled power prediction model. Furthermore, by scoring and sorting the power data of different stations, a group of stations with significant impact on the power of the target area and low similarity are determined for predicting the regional power.

[0093] In an exemplary embodiment, the above-mentioned prediction module is also used to determine multiple target measured power data corresponding to the multiple target stations and regional actual power data corresponding to the multiple target stations; determine the multiple target measured power data as model inputs and the regional actual power data as model outputs to establish a training set; wherein, there is a corresponding relationship between the target measured power data and the regional actual power data; use the training set to train the target prediction model selected from the target model database to obtain multiple prediction models; and use the multiple prediction models to predict the upscaled power of the target area.

[0094] In an exemplary embodiment, the above-mentioned prediction module is also used to identify the site type corresponding to each prediction model in the multiple prediction models; determine the prediction weight corresponding to each prediction model according to the site type; combine the multiple prediction models based on the prediction weight to obtain a combined prediction model for upscaling the power of the target area; and predict the upscaling power of the target area through the combined prediction model.

[0095] In an exemplary embodiment, the above-mentioned determination module is also used to obtain a first time series of measured power data of each site, and a second time series of regional measured power data of the area where each site is located; calculate a first correlation score between the first time series and the second time series in time series; determine a second correlation score for indicating accuracy of each site based on the measured power data and the predicted power data corresponding to the measured power data; when the installed capacity of each site is obtained, determine a third correlation score for indicating the proportion of installed capacity of each site based on the installed capacity of each site and the total installed capacity of the area corresponding to the target area; determine the evaluation score based on the first correlation score, the second correlation score, and the third correlation score.

[0096] In an exemplary embodiment, the above-mentioned determination module is also used to, when determining the first relevant score, the second relevant score, and the third relevant score corresponding to different stations, sort the different stations in different relevant dimensions according to the size of the scores to obtain the ranking values ​​corresponding to each relevant dimension of the different stations; when determining the first ranking value, the second ranking value, and the third ranking value corresponding to each station, perform weighted averaging on the first ranking value, the second ranking value, and the third ranking value to obtain the evaluation score corresponding to each station, wherein the weights corresponding to the weighted average are the first weight corresponding to the first ranking value, the second weight corresponding to the second ranking value, and the third weight corresponding to the third ranking value, respectively.

[0097] In an exemplary embodiment, the above-mentioned sorting module is also used to obtain the sorting result of the priority sorting; select the first station with the highest priority in the sorting result to add to the preset station set; calculate the geographical location similarity between the first station and the station with the highest priority among the remaining stations in the sorting result; determine other stations to be added to the station set based on the relationship between the geographical location similarity and a preset location threshold; when it is determined that the position judgment of all stations has been completed, determine a group of stations among all the stations in the station set.

[0098] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0099] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0100] S1. Obtain power data of each station in the target area, wherein the power data at least includes: measured power data of each station, predicted power data of each station, and regional measured power data of the area where each station is located;

[0101] S2. determining an evaluation score of the power data with respect to regional power data of the target area;

[0102] S3. Prioritize all stations in the target area according to the numerical values ​​of the evaluation scores, and determine a group of stations from all the stations through the ranking results, wherein the group of stations is a plurality of target stations existing in the target area, the target numerical values ​​of the plurality of target stations on the regional power are greater than a preset numerical value, and the similarity between the plurality of target stations and other stations is less than a preset similarity;

[0103] S4. Predicting the upscaled power of the target area based on the group of sites.

[0104] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0105] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0106] An embodiment of the present application further provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps in any one of the above method embodiments are performed.

[0107] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0108] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0109] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0110] S1. Obtain power data of each station in the target area, wherein the power data at least includes: measured power data of each station, predicted power data of each station, and regional measured power data of the area where each station is located;

[0111] S2. determining an evaluation score of the power data with respect to regional power data of the target area;

[0112] S3. Prioritize all stations in the target area according to the numerical values ​​of the evaluation scores, and determine a group of stations from all the stations through the ranking results, wherein the group of stations is a plurality of target stations existing in the target area, the target numerical values ​​of the plurality of target stations on the regional power are greater than a preset numerical value, and the similarity between the plurality of target stations and other stations is less than a preset similarity;

[0113] S4. Predicting the upscaled power of the target area based on the group of sites.

[0114] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0115] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0116] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0117] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for predicting regional power, characterized in that: include: Obtaining power data of each station in the target area, wherein the power data includes at least: measured power data of each station, predicted power data of each station, and regional measured power data of the area where each station is located; Determine an evaluation score of the power data with respect to regional power data of the target area; Prioritize all stations in the target area according to the numerical values ​​of the evaluation scores, and determine a group of stations among all the stations through the ranking results, wherein the group of stations is a plurality of target stations existing in the target area, the target numerical values ​​of the plurality of target stations on the regional power are greater than a preset numerical value, and the similarity between the plurality of target stations and other stations is less than a preset similarity; An upscaled power forecast is performed for the target area based on the group of sites.

2. The method according to claim 1, characterized in that The method further comprises: predicting the upscaled power of the target area based on the group of stations, comprising: Determine a plurality of target measured power data corresponding to the plurality of target stations and a plurality of regional actual power data corresponding to the plurality of target stations; Determine the plurality of target measured power data as model input and the regional actual power data as model output to establish a training set; wherein the target measured power data and the regional actual power data have a corresponding relationship; Using the training set to train the target prediction model selected from the target model database to obtain multiple prediction models; The upscaled power of the target area is predicted by using the multiple prediction models.

3. The method according to claim 2, characterized in that Predicting the upscaled power of the target area by using the multiple prediction models includes: Identifying a station type corresponding to each prediction model in the plurality of prediction models; Determine the prediction weight corresponding to each prediction model according to the station type; Combining the multiple prediction models based on the prediction weights to obtain a combined prediction model for upscaling power of the target area; The upscaled power of the target area is predicted by the combined prediction model.

4. The method according to claim 1, characterized in that Determining an evaluation score of the power data with respect to the regional power data of the target area includes: Acquire a first time series of measured power data of each of the stations and a second time series of regional measured power data of the area where each of the stations is located; Calculating a first correlation score between the first time series and the second time series in time series; Determine a second correlation score for indicating accuracy of each station according to the measured power data and predicted power data corresponding to the measured power data; When the installed capacity of each station is obtained, determining a third correlation score for indicating the installed capacity proportion of each station according to the installed capacity of each station and the total installed capacity of the region corresponding to the target region; The evaluation score is determined according to the first relevance score, the second relevance score, and the third relevance score.

5. The method according to claim 4, characterized in that Determining the evaluation score according to the first relevance score, the second relevance score, and the third relevance score includes: When the first correlation scores, the second correlation scores, and the third correlation scores corresponding to different sites are determined, the different sites are sorted in different correlation dimensions according to the scores to obtain the ranking values ​​corresponding to the different sites in each correlation dimension; When determining the first ranking value, the second ranking value, and the third ranking value corresponding to each site, a weighted average is performed on the first ranking value, the second ranking value, and the third ranking value to obtain an evaluation score corresponding to each site, wherein the weights corresponding to the weighted average are a first weight corresponding to the first ranking value, a second weight corresponding to the second ranking value, and a third weight corresponding to the third ranking value, respectively.

6. The method according to claim 1, characterized in that Prioritize all stations in the target area according to the numerical values ​​of the evaluation scores, and determine a group of stations from all the stations according to the ranking results, including: Obtaining a ranking result of the priority ranking; Select the first station with the highest priority in the sorting result to add to the preset station set; Calculating the geographical location similarity between the first station and the station with the highest priority among the remaining stations in the sorting result; Determining other stations to be added to the station set according to the relationship between the geographical location similarity and a preset location threshold; When it is determined that the positions of all the stations are determined, a group of stations from among all the stations is determined according to the station set.

7. A regional power prediction device, characterized in that: include: An acquisition module, used to acquire power data of each station in the target area, wherein the power data at least includes: measured power data of each station, predicted power data of each station, and regional measured power data of the area where each station is located; A determination module, configured to determine an evaluation score of the power data with respect to the regional power data of the target area; A sorting module, used to prioritize all stations in the target area according to the numerical values ​​of the evaluation scores, and determine a group of stations from all the stations through the sorting results, wherein the group of stations is a plurality of target stations existing in the target area, the target numerical values ​​of the plurality of target stations on the regional power are greater than a preset numerical value, and the similarity between the plurality of target stations and other stations is less than a preset similarity; A prediction module is used to predict the upscaled power of the target area based on the group of sites.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method according to any one of claims 1 to 6 when executed.

9. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 6 through the computer program.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.