CN114325877A - Method and device for evaluating weather forecast data - Google Patents
Method and device for evaluating weather forecast data Download PDFInfo
- Publication number
- CN114325877A CN114325877A CN202011062757.5A CN202011062757A CN114325877A CN 114325877 A CN114325877 A CN 114325877A CN 202011062757 A CN202011062757 A CN 202011062757A CN 114325877 A CN114325877 A CN 114325877A
- Authority
- CN
- China
- Prior art keywords
- forecast data
- forecast
- weather
- historical
- data
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000011156 evaluation Methods 0.000 claims abstract description 71
- 238000004590 computer program Methods 0.000 claims description 9
- 230000000694 effects Effects 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 7
- 238000011157 data evaluation Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 230000003203 everyday effect Effects 0.000 description 2
- 230000033001 locomotion Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000739 chaotic effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
Images
Classifications
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Provided are a method and a device for evaluating weather forecast data, wherein the evaluation method comprises the following steps: acquiring real-time weather forecast data of a wind power plant in a first preset time period; searching historical similar forecast data which is most similar to the acquired real-time weather forecast data from historical weather forecast data of the wind power plant provided by a plurality of weather sources within a first preset time period; an evaluation index for evaluating the forecast accuracy of each weather source is determined based on the forecast accuracy indicators of the historical similar forecast data. By adopting the method and the device for evaluating the weather forecast data, the forecast accuracy of each weather source can be evaluated based on the historical forecast effect of the real-time weather system, so that a more accurate forecast result of the target weather elements can be obtained.
Description
Technical Field
The present disclosure relates generally to the field of wind power generation, and more particularly, to a method and apparatus for evaluating weather forecast data.
Background
At present, data such as wind speed and wind direction of numerical weather forecast can be used as input quantity, and forecasted meteorological elements are converted into output power forecast of a wind power plant and photovoltaic by a forecasting algorithm. Therefore, accurate prediction of numerical weather forecast can provide important decision support for power scheduling, and is one of important factors for determining the prediction accuracy of the new energy power generation power.
The existing numerical prediction system can generally provide multiple predictions every day, and the prediction time efficiency can reach 5-6 days. Taking wind speed forecast as an example, the power prediction requirement of the wind power plant is to forecast every 15 minutes, namely, numerical weather forecast is required to forecast wind speed every 15 minutes.
However, at present, for wind speed prediction of a wind power plant, numerical weather forecast has the following two difficulties: firstly, the time points of the sudden rise and the sudden fall of the forecast wind speed are difficult to accurately position due to the fact that the forecast of the movement of the weather system is advanced or lagged by the numerical weather forecast. Secondly, the numerical weather forecast is called as a mesoscale numerical weather forecast, and the weather system is only forecasted on the mesoscale, so that the phenomenon of small-scale strong wind gusts at the position of the wind power plant is difficult to capture, namely, sudden changes of wind speed on the small scale cannot be accurately forecasted.
Disclosure of Invention
An object of an exemplary embodiment of the present disclosure is to provide a method and apparatus for evaluating weather forecast data, which overcome at least one of the above-mentioned disadvantages.
In one general aspect, there is provided a method of evaluating weather forecast data, the method comprising: acquiring real-time weather forecast data of a wind power plant in a first preset time period; searching historical similar forecast data which is most similar to the acquired real-time weather forecast data from historical weather forecast data of the wind power plant provided by a plurality of weather sources within a first preset time period; an evaluation index for evaluating the forecast accuracy of each weather source is determined based on the forecast accuracy indicators of the historical similar forecast data.
Optionally, the first predetermined time period comprises a plurality of time intervals, and the historical similar forecast data most similar to the real-time weather forecast data for any one of the time intervals can be determined by: and aiming at each meteorological source, acquiring historical meteorological forecast data of the meteorological source in the same time period as any time interval, and searching multiple sections of historical similar forecast data which have similarity with the real-time forecast data in any time interval from the acquired historical meteorological forecast data.
Optionally, the forecast accuracy indicator of any piece of historical similar forecast data with similarity to the real-time forecast data of any time interval searched from any one of each weather source can be determined by: acquiring historical observation data which are in the same time period with any searched segment of historical similar forecast data; determining at least one similarity index of the obtained historical observation data and any section of historical similar forecast data; and determining the prediction accuracy of any one segment of historical similar prediction data based on the at least one similar index, and taking the determined prediction accuracy as the prediction accuracy index of any one segment of historical similar prediction data.
Alternatively, an evaluation index for evaluating the accuracy of the forecast for any of the weather sources may be determined by: and determining an evaluation index for evaluating the forecast accuracy of any weather source in any time interval based on the forecast accuracy indexes of the multiple pieces of historical similar forecast data searched from any weather source.
Optionally, the step of determining an evaluation index for evaluating the forecast accuracy of said any meteorological source at said any time interval may comprise: determining an average value of the forecast accuracy indexes of the multiple pieces of historical similar forecast data as an evaluation index for evaluating the forecast accuracy of any one weather source in any one time interval, or determining the evaluation index for evaluating the forecast accuracy of any one weather source may include: determining a weight value corresponding to each section of historical similar forecast data based on the similarity degree of each section of historical similar forecast data and corresponding real-time weather forecast data, and determining an evaluation index for evaluating the forecast accuracy of any weather source in any time interval based on the forecast accuracy index of each section of historical similar forecast data and the corresponding weight value.
Optionally, the higher the similarity between any one piece of historical similar forecast data and the corresponding real-time weather forecast data is, the larger the weight value corresponding to any one piece of historical similar forecast data is.
Optionally, the evaluation method may further include: extracting a predicted value of the target meteorological element from real-time meteorological forecast data provided by each meteorological source within a second preset time period; and for each time interval contained in the second preset time period, determining a weight value corresponding to each meteorological source in the time interval according to the evaluation index of the forecasting accuracy of each meteorological source in the time interval, and obtaining a final forecasting value of the target meteorological element in the time interval based on the forecasting value of the target meteorological element in the time interval extracted from each meteorological source and the corresponding weight value.
In another general aspect, there is provided an evaluation device of weather forecast data, the evaluation device comprising: the real-time forecast data acquisition module is used for acquiring real-time weather forecast data of the wind power plant in a first preset time period; the historical similar data acquisition module is used for searching historical similar forecast data which is most similar to the acquired real-time meteorological forecast data from historical meteorological forecast data of the wind power plant provided by a plurality of meteorological sources within a first preset time period; and the forecast evaluation index determining module is used for determining an evaluation index for evaluating the forecast accuracy of each meteorological source based on the forecast accuracy index of the historical similar forecast data.
In another general aspect, there is provided a controller comprising: a processor; a memory for storing a computer program which, when executed by the processor, implements the above-described method of assessing weather forecast data.
In another general aspect, there is provided a computer readable storage medium storing a computer program which, when executed by a processor, implements the method of assessing weather forecast data as described above.
By adopting the method and the device for evaluating the weather forecast data, the forecast accuracy of each weather source can be evaluated based on the historical forecast effect of the real-time weather system, so that a more accurate forecast result of the target weather elements can be obtained.
Drawings
The above and other objects, features and advantages of the exemplary embodiments of the present disclosure will become more apparent from the following detailed description taken in conjunction with the accompanying drawings which illustrate exemplary embodiments.
FIG. 1 shows a flow diagram of a method of assessment of weather forecast data according to an exemplary embodiment of the present disclosure;
FIG. 2 shows a schematic diagram of an assessment for various meteorological sources according to an exemplary embodiment of the present disclosure;
FIG. 3 shows a flowchart of the steps of determining an evaluation index for evaluating the forecast accuracy for each meteorological source, according to an example embodiment of the present disclosure;
FIG. 4 shows a flowchart of steps to obtain a final predicted value of a target meteorological element at each time interval, according to an example embodiment of the present disclosure;
FIG. 5 shows a block diagram of an apparatus for evaluating weather forecast data according to an exemplary embodiment of the present disclosure;
fig. 6 illustrates a block diagram of a controller according to an exemplary embodiment of the present disclosure.
Detailed Description
Various example embodiments will now be described more fully with reference to the accompanying drawings, in which some example embodiments are shown.
The wind power plant is usually located in a mountainous area with complex terrain and landforms, most of the wind power plant is affected by canyon wind, and meteorological elements such as surface wind speed are difficult to forecast.
When repeating historical weather forecast data, the optimal weather source can actually report the weather process of sudden rise and sudden fall of weather elements (such as wind speed), but the existing weather source selection strategy is as follows: manually selecting according to experience, or setting an evaluation period, evaluating the meteorological element prediction accuracy of each meteorological source in the evaluation period, and automatically selecting the meteorological source with the highest accuracy. That is, based on the above weather source selection strategy, the optimal weather source cannot be selected from a plurality of weather sources in real-time weather forecast.
In order to solve the above problems, the present disclosure provides a method for evaluating the prediction accuracy of each weather source, which evaluates the prediction accuracy of each weather source based on the historical prediction effect of a real-time weather system, and can generate an optimized collective prediction result based on the evaluation result to improve the prediction accuracy of the target weather element.
Fig. 1 shows a flowchart of a method of evaluating weather forecast data according to an exemplary embodiment of the present disclosure.
Referring to FIG. 1, in step S10, real-time weather forecast data for a wind farm over a first predetermined time period is obtained.
Here, the real-time weather forecast data may be obtained by numerical weather forecast, for example, the numerical weather forecast may refer to a method of performing numerical calculation by a mainframe computer under certain initial and side conditions according to actual conditions of the atmosphere, solving a system of equations describing hydrodynamics and thermodynamics of a weather evolution process, and predicting an atmospheric motion state and a weather phenomenon in a future certain period, that is, a means of making a weather forecast using current weather conditions as input data.
In exemplary embodiments of the present disclosure, the real-time weather forecast data for the wind farm over the first predetermined time period may be real-time weather forecast data acquired from any one of a plurality of weather sources.
In step S20, the wind farm provided from the plurality of weather sources is searched for historical similar forecast data that is most similar to the acquired real-time forecast data among historical weather forecast data over a first predetermined period of time.
Here, the searched historical similar forecast data refers to historical weather forecast data that is in the same time period as the acquired real-time weather forecast data and has similarity thereto.
In an example, the real-time weather forecast data of the wind farm provided by all weather sources is obtained, and taking the station M as an example, the real-time weather forecast data of the station M in the time period from eight points earlier to five points later in the day provided by each weather source can be obtained. Here, the location of the center of the wind farm may be determined as the location of the site M.
In this case, historical weather forecast data provided by all weather sources of the wind power plant for at least 1 year are acquired, historical weather forecast data of a time period from eight hours earlier to five hours later in the day are respectively extracted from each weather source, and historical similar forecast data are searched from the extracted historical weather forecast data.
In a preferred example, the first predetermined time period may be refined, e.g., the first predetermined time period may be divided into a plurality of time intervals. In one example, the time period from eight hours earlier to five hours later may be divided into three time intervals.
Taking a time interval of 8: 00-11: 00 (hereinafter referred to as a first time interval) as an example, acquiring real-time weather forecast data of the wind power plant in the first time interval, and acquiring historical weather forecast data of the wind power plant in the first time interval for each weather source. That is, a plurality of pieces of historical weather forecast data in a first time interval may be acquired from any weather source, and similarity determination may be performed on each acquired piece of historical weather forecast data and the real-time weather forecast data, respectively, so as to determine historical weather forecast data having similarity with the real-time weather forecast data as historical similar forecast data.
Those skilled in the art can determine the historical weather forecast data with similarity to the real-time weather forecast data by using various similarity determination methods, which are not limited in the present disclosure.
In an example, historical weather forecast data having similarities to real-time weather forecast data may be determined by:
in the formula (1), E represents the correlation coefficient between the real-time weather forecast data and the historical weather forecast data, FiRepresents the ith historical weather forecast data,
representing the mean value of historical weather forecast data, OiRepresenting the ith real-time weather forecast data,
representing the average value of the real-time weather forecast data, X representing the number of historical forecast data, and Y representing the number of historical observation data.
It should be understood that the above listed similarity determination method is only an example, and the disclosure is not limited thereto.
For other time intervals, the historical similar forecast data of the real-time weather forecast data of each time interval can be obtained in the same manner as described above. By way of example, meteorological elements may include, but are not limited to, at least one of: temperature, humidity, wind speed, wind direction, atmospheric pressure at different heights.
It should be understood that the above-mentioned embodiments are only examples, and the present disclosure is not limited thereto, and those skilled in the art can adjust the specific parameters according to actual needs.
In step S30, an evaluation index for evaluating the forecast accuracy of each weather source is determined based on the forecast accuracy index of the searched historical similar forecast data.
For example, historical observation data may be acquired at the same time period as the historical similar forecast data, and forecast accuracy of the historical similar forecast data at the time period may be determined based on the acquired historical observation data.
Taking the site M as an example, after acquiring multiple pieces of historical similar forecast data in the first time interval from each weather source, respectively acquiring historical observation data corresponding to the historical weather forecast data in the first time interval, and comparing the historical observation data with the corresponding historical similar forecast data to determine the forecast accuracy of the historical similar forecast data in the first time interval.
And obtaining the evaluation index of the meteorological source in the first time interval based on the forecasting accuracy of the historical similar forecasting data of each section searched from the same meteorological source in the first time interval. That is, in the exemplary embodiment of the present disclosure, the evaluation index of each weather source at different time intervals may be obtained for different weather sources.
In the exemplary embodiment of the present disclosure, the historical observation data, the historical weather forecast data, the weather elements contained in the real-time forecast data, and the data time resolution should be consistent.
A specific process of determining an evaluation index for evaluating the forecast accuracy for each weather source is described below with reference to fig. 2 and 3.
FIG. 2 shows a schematic diagram of an assessment for various meteorological sources according to an exemplary embodiment of the present disclosure. FIG. 3 shows a flowchart of the steps of determining an evaluation index for evaluating the forecast accuracy for each meteorological source according to an example embodiment of the present disclosure.
Referring to fig. 3, in step S301, real-time weather forecast data for the ith time interval into which the first predetermined time period is divided is acquired.
For example, the first predetermined period of time may be divided at preset intervals.
In step S302, historical weather forecast data of the jth weather source in the same time period as the ith time interval is acquired, and a plurality of pieces of historical similar forecast data having similarity to the real-time forecast data of the ith time interval are searched from the acquired historical weather forecast data.
For example, taking the station M as an example, assuming that the ith time interval is the first time interval, historical weather forecast data of the jth weather source in the time interval of 8:00 to 11:00 every day can be acquired, assuming that historical weather forecast data of 360 days in the time interval of 8:00 to 11:00 are acquired, and searching multiple pieces of historical similar forecast data from the acquired historical weather forecast data of 360 days.
In step S303, historical observation data in the same period as the searched kth piece of historically similar forecast data is acquired.
For example, the kth section of historical similar forecast data is historical weather forecast data of a certain day in 360 days, and at the moment, historical observation data of the same day in a time interval of 8: 00-11: 00 can be acquired.
In step S304, at least one similarity index of the acquired historical observation data and the kth segment of historical similarity forecast data is determined.
As an example, the at least one similarity indicator may include, but is not limited to, at least one of: RMSE (root mean square error), R (correlation coefficient), MAE (mean absolute error). Here, the method for determining the above-mentioned similarity index is common knowledge in the art, and the disclosure will not be repeated for this part.
In step S305, based on at least one similarity index, a prediction accuracy of the kth segment of historical similar prediction data is determined, and the determined prediction accuracy is used as a prediction accuracy index of the kth segment of historical similar prediction data.
For example, a corresponding weight may be set for each similarity index, and the prediction accuracy of the kth segment of historical similarity prediction data is obtained based on the value of at least one similarity index and the respective corresponding weight. As an example, a weighted sum of the numerical values of the at least one similarity indicator and the respective corresponding weights may be determined as the prediction accuracy of the kth segment of historical similarity prediction data.
In an example, assuming that the wind farm assessment criterion is R, if the weight corresponding to the similar index R is set to be 1, the weights corresponding to other similar indexes are 0, and if the correlation coefficient R of the prediction of the kth segment of historical similar forecast data and the real-time weather forecast data is 0.68, the prediction accuracy of the kth segment of historical similar forecast data is 68%. Or, assuming that the wind farm assessment standard is RMSE, if the weight corresponding to the similarity index RMSE is set to 1 and the weights corresponding to other similarity indexes are set to 0, and if the root mean square error RMSE of the prediction of the kth segment of historical similar forecast data and the real-time meteorological forecast data is 2.36, the forecasting accuracy of the kth segment of historical similar forecast data is 1-1/2.36-57%. That is, the forecasting accuracy of 0% to 100% in one interval can be obtained by using the weight ratios set for different similar indexes.
In step S306, it is determined whether k is equal to P. Here, P is the number of pieces of historical similar forecast data having similarity to the real-time forecast data of the ith time interval, which are searched from the jth weather source. In this example, the initial value of k is 1.
If it is determined that k is not equal to P, step S307 is performed: so that k is k +1, and returns to at least step S303.
If it is determined that k is equal to P, step S308 is performed: and determining the evaluation index of the jth meteorological source in the ith time interval.
For example, an evaluation index for evaluating the prediction accuracy of the jth weather source in the ith time interval may be determined based on the prediction accuracy index of the multiple pieces of historical similar prediction data searched from the jth weather source.
In one example, the evaluation index may be obtained by taking an arithmetic mean.
For example, for the jth weather source, an average value of the forecast accuracy indexes of the multiple pieces of historical similar forecast data may be determined as an evaluation index for evaluating the forecast accuracy of the jth weather source in the ith time interval.
In another example, the evaluation index may be obtained by means of weighted summation.
For example, a weight value corresponding to each piece of historical similar forecast data is determined based on the similarity degree between each piece of historical similar forecast data and corresponding real-time weather forecast data, and an evaluation index for evaluating the forecast accuracy of any weather source in any time interval is determined based on the forecast accuracy index of each piece of historical similar forecast data and the corresponding weight value. Here, the higher the similarity between any piece of historical similar forecast data and the corresponding real-time weather forecast data is, the larger the weight value corresponding to the any piece of historical similar forecast data is.
For example, for the jth weather source, the forecast accuracy indicators of the multiple pieces of historical similar forecast data and various corresponding weight values may be subjected to weighted summation, and the summation value is determined as an evaluation index for evaluating the forecast accuracy of the jth weather source in the ith time interval.
In step S309, it is determined whether j is equal to n. Here, n is the number of the plurality of meteorological sources. In this example, the initial value of j is 1.
If it is determined that j is not equal to n, step S310 is performed: so that j equals j +1, and returns to execute step S302.
If it is determined that j is equal to n, step S311 is performed: it is determined whether i is equal to m. Here, m is the number of a plurality of time intervals included in the first predetermined period. In this example, the initial value of i is 1.
If it is determined that i is not equal to m, step S312 is performed: so that i becomes i +1, and returns to execute step S301.
If it is determined that i is equal to m, step S313 is performed: and obtaining the evaluation indexes of the n meteorological sources in m time intervals respectively.
Based on the mode, the evaluation indexes of different meteorological sources in a plurality of time intervals can be obtained, and the evaluation indexes can be calculated to be 0% -100% of forecasting accuracy. In an example, assuming a total of three weather sources are included, table 1 shows an example of the evaluation indices of the three weather sources at different time intervals:
TABLE 1
| Time interval | Meteorological source 1 | Meteorological source 2 | Weather source 3 |
| 2020-07-25-00: 00:00 to 2020-07-25-03: 00:00 | 56％ | 38％ | 89％ |
| 2020-07-25-03: 00:00 to 2020-07-25-06: 00:00 | 22％ | 46％ | 7％ |
| 2020-07-25_06:00:00 to 2020-07-25_09:00:00 | 78％ | 50％ | 50％ |
| 2020-07-25_09:00:00 to 2020-07-25_12:00:00 | 76％ | 93％ | 59％ |
| 2020-07-25-12: 00:00 to 2020-07-25-15: 00:00 | 34％ | 45％ | 61％ |
| 2020-07-25-15: 00:00 to 2020-07-25-18: 00:00 | 23％ | 76％ | 79％ |
| …… | …… | …… | …… |
According to the weather forecast data evaluation method, historical similar weather systems are searched for real-time forecasting according to similar criteria, an evaluation index capable of representing the real-time forecasting level of a weather source is generated by integrating the forecasting accuracy rates of a plurality of historical similar weather systems, and optimized ensemble forecasting is generated according to the evaluation index.
In an example, according to the method for evaluating weather forecast data of the exemplary embodiment of the present disclosure, after determining an evaluation index for evaluating the forecast accuracy of each weather source, the weather source may be selected to obtain the forecast value of the weather element based on the evaluation index, so as to be applied to the control of the wind farm or the photovoltaic by using the forecast value of the weather element provided by the selected weather source.
Fig. 4 shows a flowchart of steps for obtaining a final predicted value of a target meteorological element at each time interval according to an exemplary embodiment of the present disclosure.
Referring to fig. 4, in step S401, a predicted value of a target weather element is extracted from real-time weather forecast data for a second predetermined period provided from each weather source.
For example, one meteorological element may be selected as the target meteorological element from among a plurality of meteorological elements included in the real-time weather forecast data.
In step S402, the time intervals included in the second predetermined period of time, i.e., the number a of time intervals included in the second predetermined period of time, is determined.
For example, the above-mentioned first predetermined period of time is divided into a plurality of time intervals, in which case it is determined in step S402 which of the above-mentioned divided plurality of time intervals the second predetermined period of time includes.
In step S403, a weight value corresponding to each weather source in the a-th time interval is determined based on the evaluation index of the forecast accuracy of each weather source in the a-th time interval.
For example, the higher the evaluation index of the forecast accuracy of any weather source in the a-th time interval is, the larger the weight value corresponding to the any weather source in the a-th time interval is, and the lower the evaluation index of the forecast accuracy of any weather source in the a-th time interval is, the smaller the weight value corresponding to the any weather source in the a-th time interval is.
In step S404, a final predicted value of the target meteorological element in the a-th time interval is obtained based on the predicted value of the target meteorological element in the a-th time interval extracted from each meteorological source and the corresponding weight value.
For example, the predicted value of the target meteorological element in the a-th time interval extracted from each meteorological source and the corresponding weight value may be weighted and summed, and the sum value may be determined as the final predicted value of the target meteorological element in the a-th time interval.
Taking the example shown in table 1, assuming that the a-th time interval is the time interval from 2020-07-25_00:00:00 to 2020-07-25_03:00:00, the evaluation indexes of the three meteorological sources in the time interval are respectively 56%, 38% and 89%, taking the target meteorological element as the wind speed as an example, the predicted values of the target meteorological element extracted from each meteorological source in the a-th time interval are respectively v1, v2 and v3, and in this case, the final predicted value of the target meteorological element in the a-th time interval is: v-1 × 56% + v2 × 38% + v3 × 89%.
In step S405, it is determined whether a is equal to a. Here, a is the number of time intervals included in the second predetermined period. In this example, the initial value of a is 1.
If it is determined that a is not equal to A, step S406 is performed: so that a is a +1, and returns to perform step S403.
If it is determined that a is equal to A, step S407 is performed: and obtaining the final predicted value of the target meteorological element in each time interval.
According to the method for evaluating weather forecast data, all weather source forecasts are mixed into an optimized set forecast result by extracting the target weather element in each weather source forecast data and setting the weight value according to the obtained evaluation index of each weather source.
After the final predicted value of the target meteorological element in each time interval is obtained, the obtained final predicted value of the target meteorological element can be applied to control of the wind farm or the photovoltaic system.
Fig. 5 shows a block diagram of an evaluation device of weather forecast data according to an exemplary embodiment of the present disclosure.
As shown in fig. 5, the weather forecast data evaluation device 100 according to the exemplary embodiment of the present disclosure includes: a real-time forecast data acquisition module 101, a historical similar data acquisition module 102 and a forecast evaluation index determination module 103.
Specifically, the real-time forecast data acquisition module 101 acquires real-time weather forecast data of the wind farm over a first predetermined time period.
For example, the real-time weather forecast data for the wind farm over the first predetermined time period may refer to real-time weather forecast data acquired from any of a plurality of weather sources.
The historical similar data acquisition module 102 searches historical similar forecast data which is most similar to the acquired real-time forecast data from historical weather forecast data of a wind farm provided by a plurality of weather sources within a first predetermined time period.
For example, the searched historical similar forecast data may refer to historical weather forecast data that is at the same time period as, and has similarity to, the acquired real-time weather forecast data.
In an example, the historical similar data acquisition module 102 may determine the historical similar forecast data for the real-time weather forecast data for any of the time intervals by: and aiming at each meteorological source, acquiring historical meteorological forecast data of the meteorological source in the same time period as any time interval, and searching multiple sections of historical similar forecast data which are similar to the real-time forecast data in any time interval from the acquired historical meteorological forecast data.
The forecast evaluation index determination module 103 determines an evaluation index for evaluating the forecast accuracy of each weather source based on the forecast accuracy index of the searched historical similar forecast data.
For example, the forecast evaluation index determination module 103 may obtain historical observation data at the same time period as the historical similar forecast data, and determine the forecast accuracy of the historical similar forecast data at the time period based on the obtained historical observation data.
In an example, the forecast evaluation index determination module 103 may determine the forecast accuracy indicator of any piece of historical similar forecast data searched from any one of each of the weather sources that has similarity to the real-time forecast data for any time interval by: acquiring historical observation data which are in the same time period with any searched segment of historical similar forecast data; determining at least one similar index of the obtained historical observation data and any section of historical similar forecast data; and determining the prediction accuracy of any section of historical similar prediction data based on at least one similar index, and taking the determined prediction accuracy as the prediction accuracy index of any section of historical similar prediction data.
For example, the forecast evaluation index determination module 103 may determine an evaluation index for evaluating the forecast accuracy of any of the weather sources by: and determining an evaluation index for evaluating the forecast accuracy of any meteorological source in any time interval based on the forecast accuracy indexes of the multiple segments of historical similar forecast data searched from any meteorological source.
In an example, the forecast evaluation index determination module 103 may determine an average value of forecast accuracy indexes of the plurality of pieces of historical similar forecast data as an evaluation index for evaluating a forecast accuracy rate of any weather source at any time interval.
In another example, the forecast evaluation index determination module 103 may determine a weight value corresponding to each piece of historical similar forecast data based on a similarity degree of each piece of historical similar forecast data and corresponding real-time weather forecast data, and determine an evaluation index for evaluating the forecast accuracy of any weather source in any time interval based on the forecast accuracy index of each piece of historical similar forecast data and the corresponding weight value.
Here, the higher the similarity between any piece of historical similar forecast data and the corresponding real-time weather forecast data is, the larger the weight value corresponding to the any piece of historical similar forecast data is.
In an example, the weather forecast data evaluation device according to an exemplary embodiment of the present disclosure may further include: the meteorological element prediction module extracts a predicted value of the target meteorological element from real-time meteorological forecast data provided by each meteorological source in a second preset time period, determines a weight value corresponding to each meteorological source in each time interval according to an evaluation index of the forecasting accuracy of each meteorological source in the time interval aiming at each time interval contained in the second preset time period, and obtains a final predicted value of the target meteorological element in the time interval based on the predicted value of the target meteorological element in the time interval extracted from each meteorological source and the corresponding weight value.
Fig. 6 illustrates a block diagram of a controller according to an exemplary embodiment of the present disclosure.
As shown in fig. 6, the controller 200 according to an exemplary embodiment of the present disclosure includes: a processor 201 and a memory 202.
In particular, the memory 202 is used to store a computer program which, when executed by the processor 201, implements the above-described method of assessing weather forecast data.
Here, the method of evaluating weather forecast data shown in fig. 1 may be performed in the processor 201 shown in fig. 6. That is, each module shown in fig. 5 may be implemented by a general-purpose hardware processor such as a digital signal processor or a field programmable gate array, may be implemented by a special-purpose hardware processor such as a special chip, and may be implemented completely by a computer program in a software manner, for example, may be implemented as each module in the processor 201 shown in fig. 6.
There is also provided, in accordance with an exemplary embodiment of the present disclosure, a computer-readable storage medium storing a computer program. The computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to execute the above-described method of evaluating weather forecast data. The computer readable recording medium is any data storage device that can store data read by a computer system. Examples of the computer-readable recording medium include: read-only memory, random access memory, read-only optical disks, magnetic tapes, floppy disks, optical data storage devices, and carrier waves (such as data transmission through the internet via wired or wireless transmission paths).
In the existing meteorological source selection strategy, the optimal meteorological source is selected according to the recent wind speed forecasting effect, but the atmosphere is a chaotic system, and the evaluation of a single meteorological element cannot represent the forecasting effect of a meteorological source on the whole weather system at that time. In the method and the device for evaluating weather forecast data of the exemplary embodiment of the disclosure, the evaluation is performed not for a single weather element but for a weather source, so that a weather source which is more accurate for real-time weather forecast is obtained.
In the method and the device for evaluating weather forecast data of the exemplary embodiment of the disclosure, the accuracy of the real-time weather system is evaluated by searching the historical weather system similar to the real-time weather system and evaluating the forecast accuracy of the similar historical weather system, which is more scientific and reasonable compared with the existing ensemble forecast mode.
While the present disclosure has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present disclosure as defined by the following claims.
Claims (10)
1. A method for evaluating weather forecast data, the method comprising:
acquiring real-time weather forecast data of a wind power plant in a first preset time period;
searching historical similar forecast data which is most similar to the acquired real-time weather forecast data from historical weather forecast data of the wind power plant provided by a plurality of weather sources within a first preset time period;
an evaluation index for evaluating the forecast accuracy of each weather source is determined based on the forecast accuracy indicators of the historical similar forecast data.
2. The assessment method according to claim 1, wherein the first predetermined time period comprises a plurality of time intervals, and the historical similar forecast data most similar to the real-time weather forecast data is determined for any one of the time intervals by:
and aiming at each meteorological source, acquiring historical meteorological forecast data of the meteorological source in the same time period as any time interval, and searching multiple sections of historical similar forecast data which have similarity with the real-time forecast data in any time interval from the acquired historical meteorological forecast data.
3. The assessment method according to claim 2, wherein the forecast accuracy indicator of any piece of historical similar forecast data having similarity to the real-time forecast data of said any time interval searched from any one of each weather source is determined by:
acquiring historical observation data which are in the same time period with any searched segment of historical similar forecast data;
determining at least one similarity index of the obtained historical observation data and any section of historical similar forecast data;
and determining the prediction accuracy of any one segment of historical similar prediction data based on the at least one similar index, and taking the determined prediction accuracy as the prediction accuracy index of any one segment of historical similar prediction data.
4. The assessment method according to claim 2, wherein an evaluation index for evaluating the accuracy of the forecast of any one of each meteorological source is determined by:
and determining an evaluation index for evaluating the forecast accuracy of any weather source in any time interval based on the forecast accuracy indexes of the multiple pieces of historical similar forecast data searched from any weather source.
5. The assessment method according to claim 4, wherein the step of determining an evaluation index for evaluating the forecast accuracy of said any meteorological source at said any time interval comprises:
determining the average value of the forecast accuracy indexes of the multiple pieces of historical similar forecast data as an evaluation index for evaluating the forecast accuracy of any meteorological source in any time interval,
alternatively, the step of determining an evaluation index for evaluating the forecast accuracy of said any one meteorological source comprises:
determining the weight value corresponding to each section of historical similar forecast data based on the similarity degree of each section of historical similar forecast data and the corresponding real-time weather forecast data,
and determining an evaluation index for evaluating the forecasting accuracy of any meteorological source in any time interval based on the forecasting accuracy index of each section of historical similar forecasting data and the corresponding weight value.
6. The assessment method according to claim 5, wherein the higher the similarity between any piece of historical similar forecast data and the corresponding real-time weather forecast data, the higher the weight value corresponding to said any piece of historical similar forecast data.
7. The evaluation method according to claim 4, further comprising:
extracting a predicted value of the target meteorological element from real-time meteorological forecast data provided by each meteorological source within a second preset time period;
and for each time interval contained in the second preset time period, determining a weight value corresponding to each meteorological source in the time interval according to the evaluation index of the forecasting accuracy of each meteorological source in the time interval, and obtaining a final forecasting value of the target meteorological element in the time interval based on the forecasting value of the target meteorological element in the time interval extracted from each meteorological source and the corresponding weight value.
8. An apparatus for evaluating weather forecast data, the apparatus comprising:
the real-time forecast data acquisition module is used for acquiring real-time weather forecast data of the wind power plant in a first preset time period;
the historical similar data acquisition module is used for searching historical similar forecast data which is most similar to the acquired real-time meteorological forecast data from historical meteorological forecast data of the wind power plant provided by a plurality of meteorological sources within a first preset time period;
and the forecast evaluation index determining module is used for determining an evaluation index for evaluating the forecast accuracy of each meteorological source based on the forecast accuracy index of the historical similar forecast data.
9. A controller, comprising:
a processor;
a memory for storing a computer program which, when executed by the processor, implements the method of assessing weather forecast data according to any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, characterized in that the computer program, when being executed by a processor, implements the method of assessing weather forecast data according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011062757.5A CN114325877B (en) | 2020-09-30 | 2020-09-30 | Assessment method and device for weather forecast data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011062757.5A CN114325877B (en) | 2020-09-30 | 2020-09-30 | Assessment method and device for weather forecast data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114325877A true CN114325877A (en) | 2022-04-12 |
| CN114325877B CN114325877B (en) | 2024-05-17 |
Family
ID=81032004
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011062757.5A Active CN114325877B (en) | 2020-09-30 | 2020-09-30 | Assessment method and device for weather forecast data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114325877B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115629431A (en) * | 2022-12-22 | 2023-01-20 | 成都数之联科技股份有限公司 | Water vapor content prediction method, device, equipment and medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5940776A (en) * | 1996-04-12 | 1999-08-17 | Baron Services, Inc. | Automated real-time weather graphics generating systems and methods |
| US20100023266A1 (en) * | 2008-07-22 | 2010-01-28 | Henrik Stiesdal | Method and Arrangement for the Forecast of Wind-Resources |
| US20140025354A1 (en) * | 2012-07-20 | 2014-01-23 | Tata Consultancy Services Limited | Method and system for adaptive forecast of wind resources |
| CN111103635A (en) * | 2019-12-04 | 2020-05-05 | 上海眼控科技股份有限公司 | Meteorological data processing method, system, electronic equipment and storage medium |
| US20200191997A1 (en) * | 2018-12-18 | 2020-06-18 | Loon Llc | Generating weather models using real time observations |
-
2020
- 2020-09-30 CN CN202011062757.5A patent/CN114325877B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5940776A (en) * | 1996-04-12 | 1999-08-17 | Baron Services, Inc. | Automated real-time weather graphics generating systems and methods |
| US20100023266A1 (en) * | 2008-07-22 | 2010-01-28 | Henrik Stiesdal | Method and Arrangement for the Forecast of Wind-Resources |
| US20140025354A1 (en) * | 2012-07-20 | 2014-01-23 | Tata Consultancy Services Limited | Method and system for adaptive forecast of wind resources |
| US20200191997A1 (en) * | 2018-12-18 | 2020-06-18 | Loon Llc | Generating weather models using real time observations |
| CN111103635A (en) * | 2019-12-04 | 2020-05-05 | 上海眼控科技股份有限公司 | Meteorological data processing method, system, electronic equipment and storage medium |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115629431A (en) * | 2022-12-22 | 2023-01-20 | 成都数之联科技股份有限公司 | Water vapor content prediction method, device, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114325877B (en) | 2024-05-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107292098A (en) | Medium-and Long-Term Runoff Forecasting method based on early stage meteorological factor and data mining technology | |
| CN113919231B (en) | PM2.5 concentration space-time change prediction method and system based on space-time diagram neural network | |
| CN106405682B (en) | A kind of precipitation predicting method and device | |
| US20170228743A1 (en) | Crop forecasting with incremental feature selection and spectrum constrained scenario generation | |
| CN110555561A (en) | Medium-and-long-term runoff ensemble forecasting method | |
| KR101646587B1 (en) | Hybrid seasonal prediction of the Western North Pacific Tropical Cyclone Activity | |
| CN109472283B (en) | Dangerous weather prediction method and device based on multiple incremental regression tree model | |
| CN111639803A (en) | Prediction method applied to future vegetation index of area under climate change scene | |
| CN109325633A (en) | A kind of weather forecast set member choosing method and system | |
| KR20140021179A (en) | Method and apparatus for predicting daily solar radiation level | |
| CN111325376A (en) | Wind speed prediction method and device | |
| Otache et al. | ARMA modelling of Benue River flow dynamics: comparative study of PAR model | |
| CN112613633A (en) | Meteorological element prediction method and device for wind power plant | |
| CN111323847B (en) | Method and apparatus for determining weight ratios for analog integration algorithms | |
| CN115310648A (en) | Medium-and-long-term wind power combination prediction method based on multi-meteorological variable model identification | |
| CN114325877A (en) | Method and device for evaluating weather forecast data | |
| CN116976227B (en) | Storm water increasing forecasting method and system based on LSTM machine learning | |
| CN115936238B (en) | Method, system, equipment and medium for predicting medium-long-term output of global wind power | |
| CN114880373B (en) | Soil sampling method, system, storage medium and electronic equipment | |
| CN115983478A (en) | Distributed photovoltaic power generation power prediction analysis method, system, terminal and medium | |
| CN108876041B (en) | Polar sea ice area prediction method and device based on causal influence network algorithm | |
| CN113988349A (en) | Numerical model forecasting product objective release method based on space-time local model | |
| CN117195749B (en) | Regional highway network risk prediction method by climate change and readable storage medium | |
| CN115049443B (en) | Multi-scale feature-based adaptive regression meteorological medium and long term optimization method and system | |
| CN116992781B (en) | Multi-step multi-element storm forecasting method based on deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |