Method and apparatus for processing weather forecast data

Abstract

A weather forecast data processing method and a weather forecast data processing device are disclosed. The weather forecast data processing method comprises: obtaining historical weather forecast data of multiple weather sources of a wind farm and historical weather observation data of the wind farm; determining an evaluation index of each weather source based on the historical weather forecast data of the multiple weather sources and the historical weather observation data; and generating a collective weather forecast by using real-time weather forecast data of the multiple weather sources and the determined evaluation index of each weather source.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of wind power generation in general, and more particularly, to a method and device for processing meteorological forecast data suitable for wind farms. BACKGROUND

[0002] The wind speed, wind direction and other meteorological elements of numerical weather prediction can be converted into the output power prediction of wind farms and photovoltaic power plants by a prediction algorithm. Therefore, the accurate prediction of numerical weather prediction can provide important decision support for power dispatching, and is one of the important determinants of the prediction accuracy of new energy power generation.

[0003] The power prediction of a wind farm requires a prediction every 15 minutes, i.e., the numerical weather prediction is required to predict the wind speed every 15 minutes. However, there are two difficulties in the numerical weather prediction for wind speed prediction. First, the prediction of the movement of the weather system by the numerical weather prediction has a lead or lag, which makes it difficult to accurately locate the time point of sudden increase or decrease of the predicted wind speed. Second, the numerical weather prediction, also known as mesoscale numerical weather prediction, only predicts the weather system at the mesoscale, but the small-scale strong gust phenomenon at the location of the wind farm is difficult to capture. That is, the sudden change of wind speed is a difficulty in meteorological prediction.

[0004] The wind farm is usually located in a complex mountainous area, and most of them are affected by canyon wind, so it is difficult to predict the surface wind speed. By accessing the weather forecasts of multiple international authoritative agencies, ensemble prediction can be performed to reduce the power prediction accuracy caused by large wind speed prediction deviation. Currently, the problem faced by power prediction is that when reviewing the historical weather forecast, it is found that the optimal meteorological source can actually report the weather process of sudden increase or decrease of wind speed, but at that time, the optimal meteorological source cannot be selected from the many meteorological sources and used for power prediction. The current strategy for selecting a meteorological source is to select the optimal meteorological source based on the recent wind speed prediction effect, but the atmosphere itself is a chaotic system, and the evaluation of a single meteorological element cannot represent the prediction effect of the entire weather system at that time. In other words, the accuracy of evaluating a meteorological source using a single meteorological element is not high, and it is difficult to realize real-time selection of the optimal meteorological source from multiple meteorological sources. SUMMARY

[0005] Embodiments of the present disclosure provide a method and device for processing meteorological forecast data, which evaluate meteorological sources according to multiple evaluation criteria based on multiple meteorological elements rather than a single meteorological element, and generate an ensemble weather forecast using real-time meteorological forecasts of multiple meteorological sources based on the evaluation of each meteorological source, thereby significantly improving the accuracy and reliability of real-time weather forecasting.

[0006] In one general aspect, there is provided a method for processing weather forecast data, the method comprising: obtaining historical weather forecast data of a plurality of weather sources of a wind farm and historical weather observation data of the wind farm; determining an evaluation index of each weather source based on the historical weather forecast data of the plurality of weather sources and the historical weather observation data; generating an ensemble weather forecast using real-time weather forecast data of the plurality of weather sources and the determined evaluation index of each weather source.

[0007] Optionally, the step of obtaining the historical weather forecast data of the plurality of weather sources of the wind farm and the historical weather observation data of the wind farm comprises: obtaining the historical weather forecast data of the plurality of weather sources and the historical weather observation data within a predetermined time period, wherein the historical weather forecast data of each weather source comprises historical weather forecast data of a plurality of weather elements, and the historical weather observation data comprises historical weather observation data of the plurality of weather elements.

[0008] Optionally, the step of determining the evaluation index of each weather source comprises: for any one weather source, performing an accuracy evaluation on the any one weather source based on the historical weather forecast data of the plurality of weather elements of the any one weather source and the historical weather observation data of the plurality of weather elements, and determining the evaluation index of the any one weather source based on the accuracy evaluation result of the any one weather source.

[0009] Optionally, the step of performing the accuracy evaluation on the any one weather source and determining the evaluation index of the any one weather source based on the accuracy evaluation result of the any one weather source comprises: for each evaluation period in a plurality of evaluation periods, performing the following steps: performing an accuracy evaluation on the any one weather source according to a plurality of evaluation criteria based on the historical weather forecast data of the plurality of weather elements of the any one weather source and the historical weather observation data of the plurality of weather elements within the corresponding evaluation period to obtain a plurality of accuracy evaluation results; converting each accuracy evaluation result into a corresponding accuracy score, and calculating a plurality of accuracy scores as a score of the corresponding evaluation period; and determining the evaluation index of the any one weather source based on the scores of all evaluation periods.

[0010] Optionally, the predetermined time period comprises the plurality of evaluation periods.

[0011] Optionally, the plurality of evaluation criteria comprises at least one of the following: root mean square error, mean absolute error, correlation coefficient.

[0012] Optionally, the step of converting each accuracy evaluation result into a corresponding accuracy score and calculating a plurality of accuracy scores as a score of a corresponding evaluation period comprises: converting each accuracy evaluation result into a corresponding accuracy score based on an evaluation standard used to obtain each accuracy evaluation result; setting a weight for each accuracy score according to an importance of each accuracy evaluation result; and calculating the score of the corresponding evaluation period based on the plurality of accuracy scores and corresponding weights.

[0013] Optionally, the step of determining the evaluation index of the arbitrary meteorological source based on the scores of all evaluation periods comprises: determining an average of the scores of all evaluation periods as the evaluation index of the arbitrary meteorological source, or the step of determining the evaluation index of the arbitrary meteorological source based on the scores of all evaluation periods comprises: setting a weight for each evaluation period based on a time distance of each evaluation period from a real-time meteorological forecast; and determining the evaluation index of the arbitrary meteorological source based on the scores of all evaluation periods and corresponding weights.

[0014] Optionally, the step of generating a collective weather forecast using the real-time meteorological forecast data of the plurality of meteorological sources and the determined evaluation index of each meteorological source comprises: determining a weight of each meteorological source based on the evaluation index of each meteorological source; extracting a target meteorological element in the real-time meteorological forecast data of the plurality of meteorological sources; and generating the collective weather forecast based on the extracted target meteorological element and corresponding weights, wherein the real-time meteorological forecast data of each meteorological source comprises a plurality of meteorological elements.

[0015] Optionally, the historical meteorological observation data is obtained from a meteorological measuring device of a wind farm, and the plurality of meteorological elements comprises all meteorological elements or partial meteorological elements measured by the meteorological measuring device.

[0016] In another general aspect, a meteorological forecast data processing apparatus is provided, which comprises: a data acquisition unit configured to acquire historical meteorological forecast data of a plurality of meteorological sources of a wind farm and historical meteorological observation data of the wind farm; a meteorological source evaluation unit configured to determine an evaluation index of each meteorological source based on the historical meteorological forecast data of the plurality of meteorological sources and the historical meteorological observation data; and a weather forecast generation unit configured to generate a collective weather forecast using real-time meteorological forecast data of the plurality of meteorological sources and the determined evaluation index of each meteorological source.

[0017] Optionally, the data acquisition unit is configured to acquire the historical meteorological forecast data of the plurality of meteorological sources and the historical meteorological observation data within a predetermined time period, wherein the historical meteorological forecast data of each meteorological source comprises historical meteorological forecast data of a plurality of meteorological elements, and the historical meteorological observation data comprises historical meteorological observation data of the plurality of meteorological elements.

[0018] Optionally, the weather source evaluation unit is configured to: for any one of the weather sources, perform an accuracy evaluation on the any one of the weather sources based on the historical weather forecast data of the plurality of weather elements of the any one of the weather sources and the historical weather observation data of the plurality of weather elements, and determine an evaluation index of the any one of the weather sources based on the accuracy evaluation result of the any one of the weather sources.

[0019] Optionally, the weather source evaluation unit is configured to: for each of the plurality of evaluation periods, perform the following operations: based on the historical weather forecast data of the plurality of weather elements of the any one of the weather sources and the historical weather observation data of the plurality of weather elements within the corresponding evaluation period, perform an accuracy evaluation on the any one of the weather sources according to a plurality of evaluation criteria to obtain a plurality of accuracy evaluation results; convert each accuracy evaluation result into a corresponding accuracy score, and calculate a score of the corresponding evaluation period based on the plurality of accuracy scores; and determine the evaluation index of the any one of the weather sources based on the scores of all the evaluation periods.

[0020] Optionally, the predetermined time period includes the plurality of evaluation periods.

[0021] Optionally, the plurality of evaluation criteria includes at least one of the following: root mean square error, mean absolute error, correlation coefficient.

[0022] Optionally, the weather source evaluation unit is configured to: convert each accuracy evaluation result into a corresponding accuracy score based on the evaluation criteria used to obtain each accuracy evaluation result; set a weight for each accuracy score according to the importance of each accuracy evaluation result; and calculate the score of the corresponding evaluation period based on the plurality of accuracy scores and the corresponding weights.

[0023] Optionally, the weather source evaluation unit is configured to: determine the average of the scores of all the evaluation periods as the evaluation index of the any one of the weather sources, or the step of determining the evaluation index of the any one of the weather sources based on the scores of all the evaluation periods includes: setting a weight for each evaluation period based on the time distance of each evaluation period from the real-time weather forecast; and determining the evaluation index of the any one of the weather sources based on the weights of the scores of all the evaluation periods.

[0024] Optionally, the weather forecast generation unit is configured to: determine the weight of each weather source based on the evaluation index of each weather source; extract a target weather element from the real-time weather forecast data of the plurality of weather sources, wherein the real-time weather forecast data of each weather source includes a plurality of weather elements; and generate a collective weather forecast based on the extracted target weather element and the corresponding weight.

[0025] Optionally, the historical meteorological observation data is obtained from a meteorological measuring device of the wind farm, and the plurality of meteorological elements includes all or part of the meteorological elements measured by the meteorological measuring device.

[0026] In another general aspect, there is provided a computer readable storage medium storing a computer program which, when executed by a processor, implements the meteorological forecast data processing method as described above.

[0027] In another general aspect, there is provided a computing device comprising: a processor; and a memory storing a computer program which, when executed by the processor, implements the meteorological forecast data processing method as described above.

[0028] The meteorological forecast data processing method and the meteorological forecast data processing device according to embodiments of the present disclosure can improve the accuracy and reliability of real-time weather forecast, and further improve the accuracy of wind farm power prediction, based on a plurality of meteorological elements rather than a single meteorological element, evaluating meteorological sources according to a plurality of evaluation criteria, and generating a collective weather forecast using real-time weather forecasts of a plurality of meteorological sources based on the evaluation of each meteorological source.

[0029] Additional aspects and / or advantages of the general inventive concept will be set forth in part in the description that follows, and in part will be obvious from the description, or can be learned by practice of the general inventive concept. BRIEF DESCRIPTION OF DRAWINGS

[0030] The above and other objects and features of embodiments of the present disclosure will become more apparent from the following description made with reference to the accompanying drawings, in which:

[0031] Figure 1 is a flowchart illustrating a meteorological forecast data processing method according to an embodiment of the present disclosure;

[0032] Figure 2 is a flowchart illustrating a method of determining an evaluation index of a meteorological source according to an embodiment of the present disclosure;

[0033] Figure 3 is a block diagram illustrating a meteorological forecast data processing device according to an embodiment of the present disclosure;

[0034] Figure 4 is a block diagram illustrating a computing device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0035] The following detailed description is presented to aid the reader in gaining a comprehensive understanding of the methods, apparatuses, and / or systems described herein. However, various changes, modifications, and equivalents can be used, and thus particular embodiments described herein are not intended as being exhaustive of the ways in which the methods, apparatuses, and / or systems described herein can be practiced. For instance, the order in which operations are described is not intended to be limiting, except in cases where a particular order is essential, for example, where a specific sequence is required as described in the assertions presented herein. Additionally, descriptions of features in terms of other features when provided is not intended to be limiting, such descriptions are only for use in providing an overall description of embodiments disclosed herein. Furthermore, descriptions of the features in terms of their current state are provided for clarity. Descriptions of the features in terms of their future states are provided for clarity as well.

[0036] The features described herein can be implemented in different forms and should not be construed as limited to the examples described herein. Rather, these examples are provided as illustrative of only some of the many possible implementations of the methods, apparatuses, and / or systems described herein, which implementations will be readily apparent to those of ordinary skill in the art upon reading the present disclosure.

[0037] As used herein, the term "and / or" includes any one of the associated listed items, as well as any combination of any two or more of the associated listed items.

[0038] Although terms such as "first", "second", and "third" can be used herein to describe various elements, components, regions, layers or sections, these elements, components, regions, layers or sections should not be limited by these terms. Rather, these terms are only used to distinguish one element, component, region, layer or section from another element, component, region, layer or section. Thus, a first element, component, region, layer or section referred to in the examples described herein can also be called a second element, component, region, layer or section without departing from the teachings of the examples.

[0039] In the description, when an element such as a layer, a region, or a substrate is referred to as "on" another element, "connected to" or "coupled to" another element, it can be "directly on," "directly connected to," or "directly coupled to" the other element, or one or more other elements can be interposed therebetween. In contrast, when an element is referred to as being "directly on," "directly connected to," or "directly coupled to" another element, there are no other elements interposed therebetween.

[0040] The terminology used herein is for the purpose of describing various examples only and is not intended to be limiting of the disclosure. Singular forms are intended to include the plural forms, unless the context clearly indicates otherwise. The terms "comprises," "comprising," and "including" specify the presence of stated features, numbers, operations, components, elements, and / or combinations thereof, but do not preclude the presence or addition of one or more other features, numbers, operations, components, elements, and / or combinations thereof.

[0041] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. Terms such as, for example, "conventionally", "typically", and the like, are not intended to limit the scope of the present disclosure to particular examples described herein and are intended to cover various modifications and equivalents consistent with the spirit and scope of the present disclosure. Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. Unless specifically set forth herein, no term is intended to have an artificially limited meaning.

[0042] In addition, in the description of examples, detailed descriptions of related structures or functions considered to cause obscuring interpretation of the present disclosure will be omitted.

[0043] Figure 1 is a flowchart illustrating a weather forecast data processing method according to an embodiment of the present disclosure.

[0044] According to an embodiment of the present disclosure, the weather forecast data processing method according to an embodiment of the present disclosure can be performed by a wind farm master controller or a device provided in the master controller, and can also be performed by a device separately provided in a wind farm different from the master controller.

[0045] Referring to Figure 1 In step S101, historical weather forecast data of a plurality of weather sources (e.g., used weather sources) of a wind farm and historical weather observation data of the wind farm are acquired.

[0046] Specifically, in step S101, historical weather forecast data and historical weather observation data of a plurality of weather sources within a predetermined time period can be acquired. Here, the predetermined time period can be divided into a plurality of evaluation periods. In other words, the predetermined time period can include a plurality of evaluation periods. The historical weather forecast data of each weather source includes historical weather forecast data of a plurality of weather elements, and the historical weather observation data includes historical weather observation data of a plurality of weather elements. According to an embodiment of the present disclosure, the plurality of weather elements involved in the historical weather forecast data of each weather source is the same as the plurality of weather elements involved in the historical weather observation data. For example, for a wind farm M, historical weather forecast data of a plurality of weather sources within N days can be acquired, and the historical weather forecast data of each weather source can include historical weather forecast data of N weather elements (for example, but not limited to, temperature, humidity, wind speed, pressure, etc. at 10m / 30m / 50m / 70m height layer). For any weather source, the historical weather forecast data of all N weather elements at time t can be recorded as F v v i,t (i = 1, 2 …… N v ​​). On the other hand, the historical meteorological observation data of the wind farm can be obtained by a meteorological measuring device (such as but not limited to a wind tower) arranged in the wind farm, and the N v meteorological elements include all meteorological elements or part of meteorological elements measured by the meteorological measuring device. In addition, the PCA dimension reduction method can also be used to reduce the processing amount of meteorological elements. For example, for the wind farm M, the historical meteorological observation data of N v meteorological elements of the wind farm within N days can be obtained. The historical meteorological observation data of all N v meteorological elements at time t can be recorded as O i,t (i = 1, 2 …… N v ).

[0047] According to an embodiment of the present disclosure, the data time resolution of obtaining the historical meteorological forecast data of each meteorological source and the historical meteorological observation data of the wind farm can be recorded as Tmin(such as but not limited to 15 minutes). For example, the historical meteorological forecast data of multiple meteorological sources and the historical meteorological observation data of multiple meteorological sources of the wind farm M since 20:00 yesterday can be obtained at 7:00 every day. In addition, the real-time meteorological forecast data of multiple meteorological sources can be obtained at the same time, and the real-time meteorological forecast data of each meteorological source can include multiple meteorological elements.

[0048] Next, in step S102, the evaluation index of each meteorological source can be determined based on the historical meteorological forecast data and the historical meteorological observation data of multiple meteorological sources.

[0049] Specifically, for any one of the multiple meteorological sources, the accuracy of the any one of the multiple meteorological sources can be evaluated based on the historical meteorological forecast data of multiple meteorological elements of the any one of the multiple meteorological sources and the historical meteorological observation data of multiple meteorological elements, and the evaluation index of the any one of the multiple meteorological sources can be determined based on the accuracy evaluation result of the any one of the multiple meteorological sources. The following refers to Figure 2 The method for determining the evaluation index of one meteorological source is specifically described.

[0050] Figure 2 is a flow chart showing the method for determining the evaluation index of a meteorological source according to an embodiment of the present disclosure.

[0051] As described above, the predetermined time period for obtaining the historical meteorological forecast data and the historical meteorological observation data of multiple meteorological sources can be divided into multiple (for example, n) evaluation periods. The length and number of evaluation periods can be set differently according to the climate conditions of different wind farms, and the length of the default evaluation period is 7 days, and the number n = 1.

[0052] In step S201, the evaluation period count value m is set to 1. The evaluation period count value m can indicate which evaluation period the current evaluation period is.

[0053] In step S202, for a corresponding evaluation period, based on historical weather forecast data of a plurality of weather elements of a weather source in the corresponding evaluation period and historical weather observation data of the plurality of weather elements, accuracy evaluation is performed on the weather source according to a plurality of evaluation criteria to obtain a plurality of accuracy evaluation results. According to an embodiment of the present disclosure, the plurality of evaluation criteria can include at least one of root mean square error, mean absolute error, correlation coefficient. However, the present disclosure is not limited thereto, and the plurality of evaluation criteria can also include other evaluation criteria, such as mean square error, standard deviation, etc.

[0054] For example, accuracy evaluation can be performed on the weather source by the following equation (1).

[0055]

[0056] In equation (1), F t represents historical weather forecast data of all weather elements of a weather source in a corresponding evaluation period, O t may represent historical weather observation data of all weather elements of a wind farm in the corresponding evaluation period. N v represents the number of weather elements, w i represents the weight of each weather element, σ fi represents the mean absolute error of historical weather forecast data of the i-th weather element in the corresponding evaluation period, is half of the length of the corresponding evaluation period, j indicates a sampling point and increases according to the data time resolution. As the accuracy evaluation result of the weather source in the corresponding evaluation period, ‖F t ,O t ‖ can represent the degree of similarity between historical weather forecast data and historical weather observation data (i.e., the accuracy of historical weather forecast, which is essentially the root mean square error), and the smaller the value, the more accurate the historical weather forecast data. Here, it should be understood that σ fi characterizes the dispersion degree of the historical weather forecast data of the i-th weather element itself, and the purpose is to make the final accuracy evaluation result dimensionless. The weight w i of each weather element can be set according to the importance degree of each weather element to power prediction, and the present disclosure does not make special limitation thereto.

[0057] For another example, accuracy evaluation can be performed on the weather source by the following equation (2).

[0058]

[0059]

[0060]

[0061] In equation (2), Fi,t+j denotes the historical weather forecast data of the i-th weather element at time t + j (sampling point), O i,t+j denotes the historical weather observation data of the i-th weather element at time t + j (sampling point), and denote the start time and the end time of the corresponding evaluation period, respectively, and N denote the number of sampling points j within the corresponding evaluation period, σ fi denotes the mean absolute error of the historical weather forecast data of the i-th weather element within the corresponding evaluation period, denotes the mean value of the historical weather forecast data of the i-th weather element within the corresponding evaluation period, w i denotes the weight of each weather element. Similarly, as the accuracy evaluation result of the weather source within the corresponding evaluation period, ‖F t t ‖ also denotes the degree of similarity between the historical weather forecast data and the historical weather observation data (i.e., the accuracy of the historical weather forecast, which is essentially the mean absolute error), and the smaller the value, the more accurate the historical weather forecast data.

[0062] For another example, the weather source can be evaluated for accuracy by the following equation (3).

[0063]

[0064]

[0065]

[0066] In equation (3), F i,t+j denotes the historical weather forecast data of the i-th weather element at time t + j (sampling point), O i,t+j denotes the historical weather observation data of the i-th weather element at time t + j (sampling point), and denote the start time and the end time of the corresponding evaluation period, respectively, denotes the number of sampling points j within the corresponding evaluation period, denotes the mean value of the historical weather forecast data of the i-th weather element within the corresponding evaluation period, w i denotes the weight of each weather element. Similarly, as the accuracy evaluation result of the weather source within the corresponding evaluation period, ‖F t t ‖ also denotes the degree of similarity between the historical weather forecast data and the historical weather observation data (i.e., the accuracy of the historical weather forecast, which is essentially the correlation coefficient), and the larger the value, the more accurate the historical weather forecast data.

[0067] ​​The above describes several specific manners of obtaining multiple accuracy evaluation results, however, the present disclosure is not limited thereto. Any criterion capable of reflecting the similarity between historical meteorological forecast data and historical meteorological observation data can be used to evaluate the accuracy of a meteorological source, thereby obtaining a corresponding accuracy evaluation result.

[0068] Next, in step S203, each accuracy evaluation result is converted into a corresponding accuracy score, and multiple accuracy scores are calculated as a score of a corresponding evaluation period. Specifically, in step S203, each accuracy evaluation result can be first converted into a corresponding accuracy score based on the evaluation criterion used to obtain each accuracy evaluation result. Then, each accuracy score can be weighted according to the importance of each accuracy evaluation result. Finally, a score of a corresponding evaluation period is calculated based on multiple accuracy scores and corresponding weights. Here, each accuracy evaluation result can be converted into a forecast accuracy rate in an interval of 0%-100% as an accuracy score. For example, for an accuracy evaluation result obtained by equation (3) (which is essentially to calculate a correlation coefficient), if the value of the accuracy evaluation result is 0.68, it can be converted into a corresponding forecast accuracy rate (i.e., accuracy score) of 68%. For example, for an accuracy evaluation result obtained by equation (1) (which is essentially to calculate a root mean square error), if the value of the accuracy evaluation result is 2.36, it can be converted into a corresponding forecast accuracy rate (i.e., accuracy score) of 1-1 / 2.36=57%. For an accuracy evaluation result obtained by equation (2) (which is essentially to calculate a mean absolute error), a forecast accuracy rate (i.e., accuracy score) can be converted in a similar manner as equation (1). Then, by multiplying each accuracy score by a corresponding weight and summing all products, a score of a corresponding evaluation period can be calculated.

[0069] Then, in step S204, it can be determined whether the evaluation period count value m reaches the number n of evaluation periods. If m does not reach n, in step S205, m is increased by 1 (i.e., m = m + 1), and then the process returns to step S202 to calculate the score of the next evaluation period. If m reaches n, in step S206, the evaluation index of the weather source can be determined based on the scores of all the evaluation periods. Specifically, the average of the scores of all the evaluation periods can be determined as the evaluation index of the weather source. Alternatively, weights can be set for each evaluation period based on the time distance of the respective evaluation period from the real-time forecast, and then the evaluation index of the weather source can be determined based on the respective weights of the scores of all the evaluation periods. For example, assuming that the current time is September 24, the real-time weather forecast is September 25, and there are three evaluation periods, each with a length of one day, then the three evaluation periods are September 23, September 22, and September 21. According to the time distance of the three evaluation periods (September 23, September 22, and September 21) from the real-time weather forecast (September 25), the weights of the three evaluation periods (September 23, September 22, and September 21) can be set to 0.5, 0.3, and 0.2, respectively. Here, the method of setting weights for each evaluation period is exemplified, however, the present disclosure is not limited thereto, and weights can also be set for each evaluation period according to other criteria.

[0070] Referring back to Figure 1 In step S103, the ensemble weather forecast can be generated using the real-time weather forecast data of the plurality of weather sources and the determined evaluation index of each weather source.

[0071] Specifically, in step S103, the weight of each weather source can be determined based on the evaluation index of the weather source. Then, the target meteorological element in the real-time weather forecast data of the plurality of weather sources can be extracted. Finally, the ensemble weather forecast can be generated based on the extracted target meteorological element and the corresponding weight. For example, if there are five weather sources, the five weather sources can be sorted in descending order of the evaluation index. The weather source with the highest evaluation index can be assigned the largest weight, and the weather source with the lowest evaluation index can be assigned the smallest weight. If the target meteorological element is wind speed, the wind speed values can be extracted from the real-time weather forecast data of the five weather sources, then each wind speed value is multiplied by the corresponding weight, and the sum of all the products can be calculated to finally generate the ensemble weather forecast. In the present disclosure, the number of weather sources, the weight, and the target meteorological element are not limited in any way.

[0072] The weather forecast data processing method according to the embodiments of the present disclosure evaluates weather sources according to multiple evaluation criteria based on multiple weather elements instead of a single weather element, and generates a collective weather forecast by using real-time weather forecasts of multiple weather sources based on the evaluation of each weather source, which can significantly improve the accuracy and reliability of real-time weather forecasts, and further improve the accuracy of wind farm power prediction.

[0073] Figure 3 is a block diagram illustrating a weather forecast data processing apparatus according to an embodiment of the present disclosure.

[0074] Referring to Figure 3 The weather forecast data processing apparatus 300 according to the embodiments of the present disclosure can include a data acquisition unit 310, a weather source evaluation unit 320, and a weather forecast generation unit 330.

[0075] The data acquisition unit 310 can acquire historical weather forecast data of multiple weather sources of a wind farm and historical weather observation data of the wind farm. Specifically, the data acquisition unit 310 can acquire historical weather forecast data of multiple weather sources and historical weather observation data within a predetermined time period. The predetermined time period can include multiple evaluation periods. The historical weather forecast data of each weather source includes historical weather forecast data of multiple weather elements, and the historical weather observation data includes historical weather observation data of multiple weather elements. As described above, the historical weather observation data of the wind farm can be acquired by a weather measurement apparatus (for example, but not limited to, a wind measurement tower) arranged at the wind farm, and the N v weather elements include all weather elements or part of weather elements measured by the weather measurement apparatus. In addition, the data acquisition unit 310 can also acquire real-time weather forecast data of multiple weather sources, and the real-time weather forecast data of each weather source can include multiple weather elements.

[0076] The weather source evaluation unit 320 can determine an evaluation index of each weather source based on the historical weather forecast data and the historical weather observation data of multiple weather sources. Specifically, the weather source evaluation unit 320 can perform an accuracy evaluation on any one of the multiple weather sources based on the historical weather forecast data of multiple weather elements and the historical weather observation data of multiple weather elements of the any one of the multiple weather sources, and determine an evaluation index of the any one of the multiple weather sources based on the accuracy evaluation result of the any one of the multiple weather sources.

[0077] For each of the plurality of evaluation periods, the weather source evaluation unit 320 can perform accuracy evaluation on the arbitrary weather source according to the plurality of evaluation criteria based on the historical weather forecast data and the historical weather observation data of the plurality of weather elements of the arbitrary weather source in the corresponding evaluation period, to obtain a plurality of accuracy evaluation results, and then convert each accuracy evaluation result into a corresponding accuracy score, and calculate the score of the corresponding evaluation period based on the plurality of accuracy scores. The weather source evaluation unit 320 can determine the evaluation index of the arbitrary weather source based on the scores of all evaluation periods. As described above, the plurality of evaluation criteria can include at least one of, but not limited to, the root mean square error, the mean absolute error, and the correlation coefficient.

[0078] According to an embodiment of the present disclosure, the weather source evaluation unit 320 can convert each accuracy evaluation result into a corresponding accuracy score based on the evaluation criteria used to obtain each accuracy evaluation result, set a weight for each accuracy score according to the importance of each accuracy evaluation result, and calculate the score of the corresponding evaluation period based on the plurality of accuracy scores and the corresponding weights.

[0079] Alternatively, the weather source evaluation unit 320 can determine the average of the scores of all evaluation periods as the evaluation index of the arbitrary weather source. On the other hand, the weather source evaluation unit 320 can set a weight for each evaluation period based on the time distance of each evaluation period from the real-time weather forecast, and determine the evaluation index of the arbitrary weather source based on the scores of all evaluation periods and the corresponding weights.

[0080] The weather forecast generation unit 330 can generate the ensemble weather forecast by using the real-time weather forecast data of the plurality of weather sources and the determined evaluation index of each weather source. Specifically, the weather forecast generation unit 330 can determine the weight of each weather source based on the evaluation index of each weather source, extract the target weather element in the real-time weather forecast data of the plurality of weather sources, and generate the ensemble weather forecast based on the extracted target weather element and the corresponding weight.

[0081] The weather forecast data processing device according to an embodiment of the present disclosure evaluates the weather source according to the plurality of evaluation criteria based on the plurality of weather elements rather than a single weather element, and generates the ensemble weather forecast by using the real-time weather forecast of the plurality of weather sources based on the evaluation of each weather source, which can significantly improve the accuracy and reliability of the real-time weather forecast, and further improve the accuracy of the wind farm power prediction.

[0082] Figure 4 is a block diagram illustrating a computing device according to an embodiment of the present disclosure.

[0083] Referring to Figure 4The computing device 400 according to embodiments of the disclosure can include a processor 410 and a memory 420. The processor 410 can include, but is not limited to, a central processing unit (CPU), a digital signal processor (DSP), a microcomputer, a field programmable gate array (FPGA), a system on chip (SoC), a microprocessor, an application specific integrated circuit (ASIC), etc. The memory 420 stores a computer program to be executed by the processor 410. The memory 420 includes a high-speed random access memory and / or a non-volatile computer readable storage medium. When the processor 410 executes the computer program stored in the memory 420, the weather forecast data processing method as described above can be implemented.

[0084] Alternatively, the computing device 400 can communicate with various devices in the wind farm in a wired / wireless communication manner. In addition, the computing device 400 can communicate with various devices outside the wind farm in a wired / wireless communication manner.

[0085] The weather forecast data processing method according to embodiments of the disclosure can be written as a computer program and stored on a computer readable storage medium. When the computer program is executed by a processor, the weather forecast data processing method as described above can be implemented. Examples of the computer readable storage medium include a read only memory (ROM), a random access programmable read only memory (PROM), an electrically erasable programmable read only memory (EEPROM), a random access memory (RAM), a dynamic random access memory (DRAM), a static random access memory (SRAM), a flash memory, a non-volatile memory, a CD-ROM, a CD-R, a CD+R, a CD-RW, a CD+RW, a DVD-ROM, a DVD-R, a DVD+R, a DVD-RW, a DVD+RW, a DVD-RAM, a BD-ROM, a BD-R, a BD-R LTH, a BD-RE, a Blu-ray or an optical disc memory, a hard disk drive (HDD), a solid state drive (SSD), a card memory such as a multimedia card, a secure digital (SD) card or an extreme digital (XD) card, a magnetic tape, a floppy disk, a magneto-optical data storage device, an optical data storage device, a hard disk, a solid state disk, and any other device configured to store a computer program and any associated data, data files and data structures in a non-transitory manner and provide the computer program and any associated data, data files and data structures to a processor or a computer so that the processor or the computer can execute the computer program. In one example, the computer program and any associated data, data files and data structures are distributed over a networked computer system so that the computer program and any associated data, data files and data structures are stored, accessed and executed by one or more processors or computers in a distributed manner.

[0086] The weather forecast data processing method and the weather forecast data processing device according to the embodiments of the present disclosure can evaluate the weather sources according to multiple evaluation criteria based on multiple weather elements instead of a single weather element, and generate a collective weather forecast by using real-time weather forecasts of the multiple weather sources based on the evaluation of each weather source, which can significantly improve the accuracy and reliability of the real-time weather forecast, and further improve the accuracy of the wind farm power prediction.

[0087] Although some embodiments of the present disclosure have been shown and described, it should be understood by those skilled in the art that modifications can be made to these embodiments without departing from the principles and spirit of the present disclosure, which are defined by the claims and their equivalents.

Claims

1. A weather forecast data processing method characterized by comprising: The weather forecast data processing method comprises: obtaining historical weather forecast data of multiple weather sources of a wind farm and historical weather observation data of the wind farm; determining an evaluation index of each weather source based on the historical weather forecast data of the multiple weather sources and the historical weather observation data; determining a weight of each weather source based on the evaluation index of each weather source; extracting a target meteorological element in real-time weather forecast data of the multiple weather sources; generating a collective weather forecast based on the extracted target meteorological element and the corresponding weight, wherein the real-time weather forecast data of each weather source comprises multiple meteorological elements, wherein the step of determining the evaluation index of each weather source comprises, for each evaluation period in multiple evaluation periods, performing the following steps on any one weather source: based on the historical weather forecast data of the multiple meteorological elements of the any one weather source and the historical weather observation data of the multiple meteorological elements in the corresponding evaluation period, performing accuracy evaluation on the any one weather source according to multiple evaluation criteria to obtain multiple accuracy evaluation results; based on the evaluation criteria used to obtain each accuracy evaluation result, converting each accuracy evaluation result into a corresponding accuracy score; setting a weight for each accuracy score according to the importance of each accuracy evaluation result; based on the multiple accuracy scores and the corresponding weights, calculating a score of the corresponding evaluation period; determining the evaluation index of the any one weather source based on the scores of all evaluation periods.

2. The weather forecast data processing method of claim 1, wherein, The step of obtaining historical weather forecast data of multiple weather sources of a wind farm and historical weather observation data of the wind farm comprises: obtaining historical weather forecast data of the multiple weather sources and historical weather observation data in a predetermined time period, wherein the historical weather forecast data of each weather source comprises historical weather forecast data of the multiple meteorological elements, and the historical weather observation data comprises historical weather observation data of the multiple meteorological elements.

3. The weather forecast data processing method of claim 2, wherein, The predetermined time period comprises the multiple evaluation periods.

4. The weather forecast data processing method of claim 1, wherein, The multiple evaluation criteria comprise at least one of the following: root mean square error, mean absolute error, correlation coefficient.

5. The weather forecast data processing method of claim 1, wherein The step of determining the evaluation index of the any one weather source based on the scores of all evaluation periods comprises: determining the average of the scores of all evaluation periods as the evaluation index of the any one weather source, or The step of determining the evaluation index of the any one weather source based on the scores of all evaluation periods comprises: setting a weight for each evaluation period based on the time distance of each evaluation period from real-time weather forecast; determining the evaluation index of the any one weather source based on the weights of the scores of all evaluation periods.

6. The weather forecast data processing method of claim 2, wherein, The historical weather observation data is obtained from a weather measuring device of the wind farm, and the multiple meteorological elements comprise all meteorological elements or part of the meteorological elements measured by the weather measuring device.

7. A weather forecast data processing apparatus characterized by comprising: The weather forecast data processing device comprises: a data acquisition unit configured to acquire historical weather forecast data of a plurality of weather sources of a wind farm and historical weather observation data of the wind farm; a weather source evaluation unit configured to determine an evaluation index of each weather source based on the historical weather forecast data of the plurality of weather sources and the historical weather observation data; a weather forecast generation unit configured to determine a weight of each weather source based on the evaluation index of each weather source, extract a target weather element in real-time weather forecast data of the plurality of weather sources, and generate a collective weather forecast based on the extracted target weather element and the corresponding weight, wherein the real-time weather forecast data of each weather source includes a plurality of weather elements, wherein the weather source evaluation unit is configured to perform the following operations on any one weather source for each evaluation period in a plurality of evaluation periods: perform accuracy evaluation on the any one weather source according to a plurality of evaluation criteria based on historical weather forecast data of the plurality of weather elements of the any one weather source and historical weather observation data of the plurality of weather elements in a corresponding evaluation period to obtain a plurality of accuracy evaluation results; convert each accuracy evaluation result into a corresponding accuracy score based on the evaluation criteria used to obtain each accuracy evaluation result; set a weight for each accuracy score according to the importance of each accuracy evaluation result; calculate a score of the corresponding evaluation period based on the plurality of accuracy scores and the corresponding weights; determine the evaluation index of the any one weather source based on the scores of all evaluation periods.

8. A computer readable storage medium storing a computer program, characterized in that, The computer program, when executed by a processor, implements the weather forecast data processing method of any one of claims 1 to 6.

9. A computing device, comprising: The computing device includes: a processor; and a memory storing a computer program, which, when executed by the processor, implements the weather forecast data processing method of any one of claims 1 to 6.

Images