Method and device for determining wind power based on probabilistic prediction

Abstract

The application provides a wind power determination method and device based on probability prediction, which can be used in the technical field of electric power. The method comprises the following steps: for different wind resource areas, searching for a historical power prediction working condition similar to a power prediction working condition of a future moment of the wind resource area in a power prediction history database of the wind resource area; for each wind resource area, determining a power prediction error probability distribution of the wind resource area according to a wind power prediction error in the historical power prediction working condition similar to the power prediction working condition of the future moment of the wind resource area; determining a whole network power prediction error probability distribution according to the power prediction error probability distribution of each wind resource area; and correcting the predicted wind power of each wind resource area at the future moment according to the whole network prediction error probability distribution, so as to obtain the wind power of each wind resource area at the future moment. The application can improve the accuracy of wind power prediction.

Description

Technical Field

[0001] This application relates to the field of power technology, specifically to a method and apparatus for determining wind power based on probability prediction. Background Technology

[0002] With the annual growth of wind power installed capacity in the power grid, the impact of the accuracy of renewable energy forecasting on the day-ahead power balance and intraday real-time adjustments of the power grid has become increasingly significant. How to continuously improve the level of renewable energy power forecasting and better incorporate the forecast results into the day-ahead balance arrangement are two key tasks that urgently need to be carried out at this stage. Summary of the Invention

[0003] To address the problems in the prior art, embodiments of this application provide a method and apparatus for determining wind power based on probability prediction, which can at least partially solve the problems existing in the prior art.

[0004] On the one hand, this application proposes a method for determining wind power based on probabilistic prediction, including:

[0005] For different wind resource areas, search the historical power prediction database of that wind resource area for historical power prediction conditions that are similar to the power prediction conditions of that wind resource area at future times.

[0006] For each wind resource area, the probability distribution of the power prediction error for that wind resource area is determined based on the wind power prediction error under historical power prediction conditions that are similar to the power prediction conditions for future times in that wind resource area.

[0007] Based on the power prediction error probability distribution of each wind resource area, determine the power prediction error probability distribution of the entire network;

[0008] Based on the probability distribution of the prediction error across the entire network, the predicted wind power for each wind resource region at the future time is corrected to obtain the wind power for each wind resource region at the future time.

[0009] In some embodiments, the method further includes:

[0010] The day-ahead power generation plan of the power grid is determined based on the wind power output of each wind resource area at the future time.

[0011] In some embodiments, searching the historical power forecast database of different wind resource areas for historical power forecast conditions similar to the future power forecast conditions of the wind resource area includes:

[0012] For different wind resource areas, obtain the values ​​of the influence factors for each historical power prediction condition in the historical power prediction database of that wind resource area;

[0013] If the value of the influence factor of the historical power prediction operating condition and the value of the influence factor of the power prediction operating condition at a future time meet the preset constraints, then it is confirmed that the historical power prediction operating condition and the power prediction operating condition at a future time are similar.

[0014] In some embodiments, the influencing factors include at least two;

[0015] The step of confirming that the historical power prediction operating condition and the future power prediction operating condition are similar if the values ​​of the influence factors of the historical power prediction operating condition and the future power prediction operating condition meet the preset constraints includes:

[0016] If the values ​​of each influencing factor under the historical power prediction conditions meet the corresponding preset constraints with the values ​​of the corresponding influencing factors under the power prediction conditions at future times, then the historical power prediction conditions are confirmed to be similar to the current power prediction conditions.

[0017] In some embodiments, the influencing factors include the ratio of the predicted power at the prediction time to the wind power installed capacity of the wind resource area at the prediction time, the ratio of the predicted power at the adjacent time before the prediction time to the wind power installed capacity of the wind resource area at the adjacent time, and the ratio of the predicted power at the adjacent time after the prediction time to the wind power installed capacity of the wind resource area at the adjacent time.

[0018] In some embodiments, determining the power prediction error probability distribution for each wind resource area based on the wind power prediction error under historical power prediction conditions similar to the future power prediction conditions of that wind resource area includes:

[0019] For each wind resource area, the probability distribution of power prediction error for that wind resource area is determined by using the kernel density function method based on the wind power prediction error under historical power prediction conditions similar to the future power prediction conditions of that wind resource area.

[0020] In some embodiments, determining the probability distribution of power prediction errors for the entire network based on the probability distribution of power prediction errors for each wind resource region includes:

[0021] Based on the power prediction error probability distribution of each wind resource area, the Monte Carlo sampling method is used to fit the power prediction error probability distribution of the entire network.

[0022] In some embodiments, the step of correcting the predicted wind power of each wind resource area at the future time based on the probability distribution of the prediction error of the entire network, to obtain the wind power of each wind resource area at the future time, includes:

[0023] Based on the probability distribution of the prediction error of the entire network, the average prediction error, upper limit of bandwidth, and lower limit of bandwidth of the wind power of the entire network are determined.

[0024] During peak load periods, the predicted wind power of each wind resource area is corrected using the lower limit of the bandwidth; during off-peak load periods, the predicted wind power of each wind resource area is corrected using the upper limit of the bandwidth; and at other times, the predicted wind power of each wind resource area is corrected according to the average prediction error, thus obtaining the wind power of each wind resource area at the future time.

[0025] On the other hand, this application proposes a wind power determination device based on probability prediction, comprising:

[0026] The search module is used to search the historical power prediction database of different wind resource areas for historical power prediction conditions that are similar to the power prediction conditions of the wind resource area at future times.

[0027] The first determining module is used to determine the probability distribution of power prediction error for each wind resource area based on the wind power prediction error under historical power prediction conditions similar to the future power prediction conditions of the wind resource area.

[0028] The second determining module is used to determine the probability distribution of power prediction error for the entire network based on the probability distribution of power prediction error for each wind resource area.

[0029] The correction module is used to correct the predicted wind power of each wind resource area at the future time according to the probability distribution of the prediction error of the whole network, so as to obtain the wind power of each wind resource area at the future time.

[0030] In some embodiments, the apparatus further includes:

[0031] The third determining module is used to determine the day-ahead power generation plan of the power grid based on the wind power output of each wind resource area at the future time.

[0032] In some embodiments, the lookup module includes:

[0033] The acquisition unit is used to acquire the values ​​of the influence factors for each historical power prediction condition in the historical power prediction database of different wind resource areas.

[0034] The determining unit is configured to confirm that the historical power prediction operating condition is similar to the future power prediction operating condition if the value of the influence factor of the historical power prediction operating condition and the value of the influence factor of the future power prediction operating condition meet the preset constraints.

[0035] In some embodiments, the influencing factors include at least two types; the determining unit is specifically used for:

[0036] If the values ​​of each influencing factor under the historical power prediction conditions meet the corresponding preset constraints with the values ​​of the corresponding influencing factors under the power prediction conditions at future times, then the historical power prediction conditions are confirmed to be similar to the current power prediction conditions.

[0037] In some embodiments, the influencing factors include the ratio of the predicted power at the prediction time to the wind power installed capacity of the wind resource area at the prediction time, the ratio of the predicted power at the adjacent time before the prediction time to the wind power installed capacity of the wind resource area at the adjacent time, and the ratio of the predicted power at the adjacent time after the prediction time to the wind power installed capacity of the wind resource area at the adjacent time.

[0038] In some embodiments, the first determining module is specifically used for:

[0039] For each wind resource area, the probability distribution of power prediction error for that wind resource area is determined by using the kernel density function method based on the wind power prediction error under historical power prediction conditions similar to the future power prediction conditions of that wind resource area.

[0040] In some embodiments, the second determining module is specifically used for:

[0041] Based on the power prediction error probability distribution of each wind resource area, the Monte Carlo sampling method is used to fit the power prediction error probability distribution of the entire network.

[0042] In some embodiments, the correction module is specifically used for:

[0043] Based on the probability distribution of the prediction error of the entire network, the average prediction error, upper limit of bandwidth, and lower limit of bandwidth of the wind power of the entire network are determined.

[0044] During peak load periods, the predicted wind power of each wind resource area is corrected using the lower limit of the bandwidth; during off-peak load periods, the predicted wind power of each wind resource area is corrected using the upper limit of the bandwidth; and at other times, the predicted wind power of each wind resource area is corrected according to the average prediction error, thus obtaining the wind power of each wind resource area at the future time.

[0045] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of the wind power determination method based on probability prediction described in any of the above embodiments.

[0046] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the wind power determination method based on probability prediction described in any of the above embodiments.

[0047] The wind power prediction method and apparatus provided in this application collect wind farm data from the entire network and establish historical power prediction databases for different wind resource areas. It uses a working condition identification method to determine the probability distribution of wind power prediction errors in each wind resource area, and further determines the probability distribution of prediction errors across the entire network based on the probability distribution of wind power prediction errors in each wind resource area. Finally, it corrects the predicted power based on the probability distribution of prediction errors across the entire network, thereby improving the accuracy of wind power prediction. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0049] Figure 1 This is a flowchart illustrating a wind power determination method based on probability prediction provided in an embodiment of this application.

[0050] Figure 2 This is a partial flowchart of a wind power determination method based on probability prediction provided in an embodiment of this application.

[0051] Figure 3 This is a flowchart illustrating a wind power determination method based on probability prediction provided in an embodiment of this application.

[0052] Figure 4 This is a probability distribution map of prediction error for a certain wind resource area provided in an embodiment of this application.

[0053] Figure 5 This is a probability distribution diagram of the prediction error of a power grid at a certain moment in March 2018, provided by an embodiment of this application.

[0054] Figure 6 This is a partial flowchart of a wind power determination method based on probability prediction provided in an embodiment of this application.

[0055] Figure 7 This is a flowchart illustrating a wind power determination method based on probability prediction provided in an embodiment of this application.

[0056] Figure 8This is a schematic diagram of the structure of a wind power determination device based on probability prediction provided in an embodiment of this application.

[0057] Figure 9 This is a schematic diagram of the physical structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the embodiments of this application will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments and their descriptions are used to explain this application, but are not intended to limit this application. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be arbitrarily arranged.

[0059] The terms “first,” “second,” etc., used in this document are not intended to specifically refer to order or sequence, nor are they used to limit this application; they are merely used to distinguish elements or operations described using the same technical terms.

[0060] The terms “include,” “including,” “have,” “contain,” etc., used in this article are all open-ended terms, meaning that they include but are not limited to.

[0061] The term "and / or" as used in this document includes any or all of the items mentioned.

[0062] To better understand this application, the research background of this application will be explained below.

[0063] Currently, power grid dispatching agencies mainly use the revised short-term power forecast results (point forecasts) reported by wind farms to incorporate into the day-ahead power balance plan. Due to the low accuracy of numerical weather forecasts for wind farms, the short-term wind power forecast error is relatively large. Therefore, it is necessary to estimate the uncertainty of wind power forecasts so that the short-term wind power forecasts can be reasonably corrected before being incorporated into the day-ahead balance arrangement.

[0064] Wind power probabilistic prediction refers to establishing a predictive model based on meteorological data, historical wind power prediction data, and measured data to predict the fluctuation range and distribution function of future wind power, addressing the uncertainty of wind farm power. Researchers have conducted extensive research on wind power probabilistic prediction. Based on whether the wind power error distribution follows a known distribution, wind power probabilistic prediction can be divided into parametric modeling and non-parametric modeling. Based on whether it assumes the power error distribution is related to other input variables, it can be divided into conditional modeling and unconditional modeling. Researchers have conducted relevant research on all these modeling methods. In practical applications, some wind power prediction software already includes probabilistic prediction functions. Due to significant differences in wind resource characteristics across regions and varying levels of wind farm technology and management, it is necessary to propose wind power probabilistic prediction methods suitable for actual grid application conditions and incorporate the prediction results into day-ahead power balance optimization. For wind power probabilistic prediction, the prediction error characteristics will differ depending on the prediction operating conditions. Predicted power and meteorological factors are all factors that affect the wind power prediction conditions. For example, since wind power is proportional to the cube of wind speed, a small wind speed prediction error will cause a large wind power prediction error in the steep section of the wind power curve.

[0065] The execution subject of the wind power determination method based on probability prediction provided in the embodiments of this application includes, but is not limited to, a computer.

[0066] Figure 1 This is a flowchart illustrating a wind power determination method based on probabilistic prediction provided in an embodiment of this application, as shown below. Figure 1 As shown in the embodiments of this application, the wind power determination method based on probability prediction includes:

[0067] S101. For different wind resource areas, search the historical power prediction database of the wind resource area for historical power prediction conditions that are similar to the power prediction conditions of the wind resource area at future times.

[0068] S102. For each wind resource area, determine the probability distribution of the power prediction error of the wind resource area based on the wind power prediction error under historical power prediction conditions similar to the power prediction conditions of the wind resource area at future times.

[0069] S103. Determine the probability distribution of power prediction error for the entire network based on the probability distribution of power prediction error for each wind resource area.

[0070] S104. Based on the probability distribution of the prediction error of the entire network, the predicted wind power of each wind resource area at the future time is corrected to obtain the wind power of each wind resource area at the future time.

[0071] Specifically, wind speed and power of wind farms or wind turbines are not statistically independent in space, but rather correlated. Therefore, the predicted and actual power of wind farms in the same wind resource area have a certain spatial correlation. Wind farms in the same wind resource area typically transmit wind power through centralized substations, and their actual power is related to the transmission capacity of the lines. Since wind power and installed capacity are correlated, historical power prediction databases can also record the installed wind power capacity of wind resource areas at different times. This application establishes historical power prediction databases based on wind resource areas. Each database can record the historical power prediction conditions of each wind resource area, such as historical predicted power and historical installed wind power capacity.

[0072] Currently, most wind farms use short-term power forecasting products that primarily obtain forecast results by inputting numerical weather predictions provided by meteorological agencies into wind power conversion models. Because current weather forecasting technology cannot provide completely accurate, location-specific, and quantitative forecasts of wind processes, numerical weather predictions contain errors in predicting the phase and amplitude of strong wind events. Furthermore, since power and wind speed have a cubic relationship, wind power conversion models amplify these numerical weather prediction errors. Grid dispatching agencies incorporate wind power forecast results into their day-ahead power balancing plans; therefore, it is necessary to describe these uncertainties to assist grid decision-making. This application, based on a historical power forecasting database, obtains historical power forecasting conditions similar to those at a specified future time within the same wind resource region.

[0073] This application proposes a wind power prediction method based on probability forecasting. This method collects wind farm data from the entire network and establishes historical power prediction databases for different wind resource areas. It then uses a condition identification method to determine the probability distribution of wind power prediction errors for each wind resource area, and further determines the probability distribution of prediction errors for the entire network based on this distribution. Finally, it corrects the predicted power based on the probability distribution of prediction errors across the entire network, thereby improving the accuracy of wind power prediction.

[0074] In some embodiments, the method further includes: determining a day-ahead power generation plan for the power grid based on the wind power output of each wind resource area at the future time.

[0075] Specifically, the wind power output of each wind resource area at the future time will be incorporated into the day-ahead power generation plan of the power grid, thus ensuring the safety and economy of the power system.

[0076] like Figure 2 As shown, in some embodiments, the step of searching the historical power prediction database of different wind resource areas for historical power prediction conditions similar to the future power prediction conditions of the wind resource area includes:

[0077] S1011. For different wind resource areas, obtain the values ​​of the influence factors for each historical power prediction condition in the historical power prediction database of that wind resource area.

[0078] S1012. If the value of the influence factor of the historical power prediction operating condition and the value of the influence factor of the power prediction operating condition at a future time meet the preset constraint conditions, then it is confirmed that the historical power prediction operating condition and the power prediction operating condition at a future time are similar.

[0079] Specifically, factors such as predicted power and meteorological conditions can be used as influencing factors to identify wind power prediction conditions, and similar conditions can be found in historical databases.

[0080] In some embodiments, the influencing factors include at least two types; the step of confirming that the historical power prediction operating condition and the future power prediction operating condition are similar if the values ​​of the influencing factors of the historical power prediction operating condition and the values ​​of the influencing factors of the future power prediction operating condition meet preset constraints includes:

[0081] If the values ​​of each influencing factor under the historical power prediction conditions meet the corresponding preset constraints with the values ​​of the corresponding influencing factors under the power prediction conditions at future times, then the historical power prediction conditions are confirmed to be similar to the current power prediction conditions.

[0082] For example, let K be a set of wind resource regions, k∈K. Let x t,k i Let y represent the i-th influencing factor of the power prediction condition of wind resource region k at time t. t,k Let represent the predicted power value of wind resource region k at time t. Here, i ∈ I, and I is the set of influencing factors. Assume that each influencing factor is bounded and can be homogenized. There is x t,k i ∈X i , where X i Let be the range of values ​​for the i-th influence factor.

[0083] Power prediction condition c for wind resource area k at time t t,k It can be represented by the set of k-influence factors of wind resource region at that moment:

[0084] c t,k ={x t,k 1 ,x t,k 2 ,...,x t,k I},c t,k ∈C k =X 1×X 2 ×...×X I ;

[0085] Where C k It is the set of predicted operating conditions for wind resource region k.

[0086] When the wind resource area k is predicted to have power under condition c at time t in the future. t,k All influencing factors and their historical predicted working conditions c at time t0 t0,k The corresponding influence factor values ​​are close, that is... have When the predicted wind resource area k is close to the predicted operating conditions at the historical time t0, the prediction error distribution has similar characteristics, and the prediction error distribution at time t0 can be estimated based on the prediction at time t0.

[0087] In some embodiments, the influencing factors include the ratio of the predicted power at the prediction time to the wind power installed capacity of the wind resource area at the prediction time, the ratio of the predicted power at the adjacent time before the prediction time to the wind power installed capacity of the wind resource area at the adjacent time, and the ratio of the predicted power at the adjacent time after the prediction time to the wind power installed capacity of the wind resource area at the adjacent time.

[0088] Specifically, due to limitations in management and technology, wind farms can currently only ensure the accuracy and reliability of power data reported to the grid dispatching agency. Reporting of other data, such as wind speed and individual turbine information, is not yet standardized. Therefore, in this example, predicted power is used as the influencing factor for the predicted operating condition. Simultaneously, considering the time delay of wind speed and the accurate identification of wind power ramp-up, the predicted power for a period before and after time t should also be included as an influencing factor for the predicted operating condition. Finally, since predicted power is positively correlated with the total installed capacity of the wind resource area, it is necessary to standardize the predicted power (the proportion of predicted power to the total installed capacity of the wind resource area) to eliminate interference from changes in installed capacity on the identification of similar operating conditions.

[0089] In some embodiments, determining the power prediction error probability distribution for each wind resource area based on the wind power prediction error under historical power prediction conditions similar to the future power prediction conditions of that wind resource area includes:

[0090] For each wind resource area, the probability distribution of power prediction error for that wind resource area is determined by using the kernel density function method based on the wind power prediction error under historical power prediction conditions similar to the future power prediction conditions of that wind resource area.

[0091] Specifically, such as Figure 3As shown, the kernel density function method is used to fit the set of historical prediction errors with similar prediction conditions to an error probability distribution function. Probability density estimation estimates the probability density function of the population from a sample, and is usually divided into parametric and non-parametric forms. Kernel density estimation is a commonly used and effective non-parametric probability density estimation method, and its expression is shown below:

[0092]

[0093] In the formula, n is the number of samples; h is the bandwidth; x i is the sample; K(﹒) is the kernel function.

[0094] This application uses the Gaussian kernel function, the expression of which is shown below:

[0095]

[0096] The selection of the kernel density estimation bandwidth h affects the estimation accuracy. When the bandwidth is small, the probability density curve is too sharp, and the results fluctuate greatly; when the bandwidth is large, the probability density curve is too smooth, and the true nature of the data may be masked. The bandwidth can be set according to the specific use case.

[0097] This application selects a set of historical prediction times with similar prediction conditions to the future power prediction time of a wind resource area. Using the power prediction errors corresponding to these historical prediction times as samples, a Gaussian kernel function is used to estimate the probability density, generating a prediction error probability distribution. For example, the prediction error probability distribution for a certain wind resource area is as follows: Figure 4 As shown (unit: MW).

[0098] In some embodiments, determining the probability distribution of power prediction errors for the entire network based on the probability distribution of power prediction errors for each wind resource region includes:

[0099] Based on the power prediction error probability distribution of each wind resource area, the Monte Carlo sampling method is used to fit the power prediction error probability distribution of the entire network.

[0100] Specifically, since the probability distribution of future power prediction errors for a single wind resource area obtained using the operating condition identification and kernel density estimation methods is not approximately normal or other regular distributions, exhibiting significant kurtosis and skewness, this application employs a scenario sampling method for irregular probability distributions. After obtaining the probability distribution of power prediction errors for each wind resource area of ​​the power grid, the distribution is converted according to the installed capacity of each area and then sampled using the Monte Carlo method. The sampling results for each area are summed, and after summing a large number of sampling results, Monte Carlo sampling is performed again. The probability distribution of wind power prediction for the entire network is obtained by fitting the interval distribution of the sampling results. For example, the probability distribution of prediction errors for the entire power grid at a certain moment in March 2018 (unit: MW) is as follows: Figure 5As shown.

[0101] like Figure 6 As shown, in some embodiments, the step of correcting the predicted wind power of each wind resource area at the future time based on the probability distribution of the prediction error of the entire network, to obtain the wind power of each wind resource area at the future time, includes:

[0102] S1041. Based on the probability distribution of the prediction error of the entire network, determine the average prediction error of the wind power of the entire network, the upper limit of the bandwidth, and the lower limit of the bandwidth.

[0103] S1042. During peak load periods, the predicted wind power of each wind resource area is corrected using the lower limit of the bandwidth; during off-peak load periods, the predicted wind power of each wind resource area is corrected using the upper limit of the bandwidth; and at other times, the predicted wind power of each wind resource area is corrected according to the average prediction error, so as to obtain the wind power of each wind resource area at the future time.

[0104] Specifically, the wind power prediction error equals the predicted output minus the actual output. For example... Figure 7 As shown, when the prediction error is positive and the predicted output is greater than the actual output, the grid's reserve capacity needs to be used to fill the power shortfall; when the prediction error is negative and the predicted output is less than the actual output, the grid's peak-shaving margin needs to be used to absorb the excess wind power. During peak load periods, thermal power units have higher output and the grid's reserve capacity is lower; conversely, during off-peak load periods, thermal power units have lower output and the peak-shaving margin is higher. During peak load periods, positive prediction errors in wind power should be minimized as much as possible, and during off-peak load periods, negative prediction errors should be minimized as much as possible. Therefore, when incorporating wind power predictions into the day-ahead plan, the lower limit of the bandwidth (maximum negative prediction error) should be used during peak load periods to avoid positive or negative prediction errors, and the upper limit of the bandwidth (maximum positive prediction error) should be used during off-peak load periods to avoid negative prediction errors. For other times, the average error should be used for correction.

[0105] To address the uncertainty in wind power forecasting, the method proposed in this application can estimate the probability distribution of errors by combining historical data. Different strategies are employed during different load periods to incorporate the forecasted power into the day-ahead power balance plan. Simulation results verify the effectiveness of the method. The proposed method can meet the power balance requirements of power systems including wind power. Compared with traditional wind power participation methods, it can reduce redundant thermal power capacity during peak hours and spinning reserve during off-peak hours, ensuring power system security and economy, and contributing to a high proportion of wind power consumption.

[0106] Figure 8 This is a schematic diagram of a wind power determination device based on probability prediction provided in an embodiment of this application, as shown below. Figure 8As shown in the embodiment of this application, the wind power determination device based on probability prediction includes:

[0107] The search module 21 is used to search for historical power prediction conditions similar to the future power prediction conditions of different wind resource areas in the historical power prediction database of the wind resource area.

[0108] The first determining module 22 is used to determine the probability distribution of power prediction error for each wind resource area based on the wind power prediction error under historical power prediction conditions similar to the power prediction conditions of the wind resource area at future times.

[0109] The second determining module 23 is used to determine the probability distribution of power prediction error for the entire network based on the probability distribution of power prediction error for each wind resource area.

[0110] The correction module 24 is used to correct the predicted wind power of each wind resource area at the future time according to the probability distribution of the prediction error of the whole network, so as to obtain the wind power of each wind resource area at the future time.

[0111] This application proposes a wind power prediction-based device that collects wind farm data from the entire network and establishes historical power prediction databases for different wind resource areas. It then uses a condition identification method to determine the probability distribution of wind power prediction errors for each wind resource area, and further determines the probability distribution of prediction errors for the entire network based on this distribution. Finally, it corrects the predicted power based on the probability distribution of prediction errors across the entire network, thereby improving the accuracy of wind power prediction.

[0112] In some embodiments, the apparatus further includes:

[0113] The third determining module is used to determine the day-ahead power generation plan of the power grid based on the wind power output of each wind resource area at the future time.

[0114] In some embodiments, the lookup module includes:

[0115] The acquisition unit is used to acquire the values ​​of the influence factors for each historical power prediction condition in the historical power prediction database of different wind resource areas.

[0116] The determining unit is configured to confirm that the historical power prediction operating condition is similar to the future power prediction operating condition if the value of the influence factor of the historical power prediction operating condition and the value of the influence factor of the future power prediction operating condition meet the preset constraints.

[0117] In some embodiments, the influencing factors include at least two types; the determining unit is specifically used for:

[0118] If the values ​​of each influencing factor under the historical power prediction conditions meet the corresponding preset constraints with the values ​​of the corresponding influencing factors under the power prediction conditions at future times, then the historical power prediction conditions are confirmed to be similar to the current power prediction conditions.

[0119] In some embodiments, the influencing factors include the ratio of the predicted power at the prediction time to the wind power installed capacity of the wind resource area at the prediction time, the ratio of the predicted power at the adjacent time before the prediction time to the wind power installed capacity of the wind resource area at the adjacent time, and the ratio of the predicted power at the adjacent time after the prediction time to the wind power installed capacity of the wind resource area at the adjacent time.

[0120] In some embodiments, the first determining module is specifically used for:

[0121] For each wind resource area, the probability distribution of power prediction error for that wind resource area is determined by using the kernel density function method based on the wind power prediction error under historical power prediction conditions similar to the future power prediction conditions of that wind resource area.

[0122] In some embodiments, the second determining module is specifically used for:

[0123] Based on the power prediction error probability distribution of each wind resource area, the Monte Carlo sampling method is used to fit the power prediction error probability distribution of the entire network.

[0124] In some embodiments, the correction module is specifically used for:

[0125] Based on the probability distribution of the prediction error of the entire network, the average prediction error, upper limit of bandwidth, and lower limit of bandwidth of the wind power of the entire network are determined.

[0126] During peak load periods, the predicted wind power of each wind resource area is corrected using the lower limit of the bandwidth; during off-peak load periods, the predicted wind power of each wind resource area is corrected using the upper limit of the bandwidth; and at other times, the predicted wind power of each wind resource area is corrected according to the average prediction error, thus obtaining the wind power of each wind resource area at the future time.

[0127] The embodiments of the apparatus provided in this application can be used to execute the processing flow of the above-described method embodiments. Its functions will not be repeated here, but can be referred to the detailed description of the above-described method embodiments.

[0128] Figure 9 This is a schematic diagram of the physical structure of an electronic device provided in an embodiment of this application, as shown below. Figure 9 As shown, the electronic device may include a processor 301, a communications interface 302, a memory 303, and a communication bus 304, wherein the processor 301, the communications interface 302, and the memory 303 communicate with each other via the communication bus 304. The processor 301 may call logical instructions in the memory 303 to execute the methods described in any of the above embodiments.

[0129] Furthermore, the logical instructions in the aforementioned memory 303 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0130] This embodiment discloses a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, and when the program instructions are executed by a computer, the computer can perform the methods provided in the above-described method embodiments.

[0131] This embodiment provides a computer-readable storage medium storing a computer program that causes the computer to perform the methods provided in the above-described method embodiments.

[0132] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0133] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0134] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0135] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0136] In the description of this specification, the references to terms such as "an embodiment," "a specific embodiment," "some embodiments," "for example," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0137] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above descriptions are merely specific embodiments of this application and are not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for determining wind power output based on probabilistic prediction, characterized in that, include: For different wind resource areas, search the historical power prediction database of that wind resource area for historical power prediction conditions that are similar to the power prediction conditions of that wind resource area at future times. For each wind resource area, the probability distribution of the power prediction error for that wind resource area is determined based on the wind power prediction error under historical power prediction conditions that are similar to the power prediction conditions for future times in that wind resource area. Based on the power prediction error probability distribution of each wind resource area, the Monte Carlo sampling method is used to fit the power prediction error probability distribution of the entire network. Based on the probability distribution of the power prediction error of the entire network, the predicted wind power of each wind resource area at the future time is corrected to obtain the wind power of each wind resource area at the future time. Specifically, for different wind resource areas, searching the historical power prediction database for that wind resource area for historical power prediction conditions similar to the future power prediction conditions of that wind resource area includes: For different wind resource areas, obtain the values ​​of the influence factors for each historical power prediction condition in the historical power prediction database of that wind resource area; If the value of the influence factor of the historical power prediction operating condition and the value of the influence factor of the power prediction operating condition at a future time meet the preset constraint conditions, then it is confirmed that the historical power prediction operating condition and the power prediction operating condition at a future time are similar. The influencing factors include the ratio of the predicted power at the prediction time to the wind power installed capacity of the wind resource area at that prediction time, the ratio of the predicted power at the adjacent time before the prediction time to the wind power installed capacity of the wind resource area at that adjacent time, and the ratio of the predicted power at the adjacent time after the prediction time to the wind power installed capacity of the wind resource area at that adjacent time.

2. The method according to claim 1, characterized in that, The method further includes: The day-ahead power generation plan of the power grid is determined based on the wind power output of each wind resource area at the future time.

3. The method according to claim 1, characterized in that, The influencing factors include at least two types; The step of confirming that the historical power prediction operating condition and the future power prediction operating condition are similar if the values ​​of the influence factors of the historical power prediction operating condition and the future power prediction operating condition meet the preset constraints includes: If the values ​​of each influencing factor under the historical power prediction conditions meet the corresponding preset constraints with the values ​​of the corresponding influencing factors under the power prediction conditions at future times, then the historical power prediction conditions are confirmed to be similar to the current power prediction conditions.

4. The method according to claim 1, characterized in that, For each wind resource area, determining the probability distribution of the power prediction error for that wind resource area based on the wind power prediction error under historical power prediction conditions similar to the future power prediction conditions for that wind resource area includes: For each wind resource area, the probability distribution of power prediction error for that wind resource area is determined by using the kernel density function method based on the wind power prediction error under historical power prediction conditions similar to the future power prediction conditions of that wind resource area.

5. The method according to claim 1, characterized in that, The step of correcting the predicted wind power of each wind resource area at the future time based on the probability distribution of the power prediction error of the entire network, to obtain the wind power of each wind resource area at the future time, includes: Based on the probability distribution of the power prediction error of the entire network, the average prediction error, upper limit of bandwidth, and lower limit of bandwidth of the wind power of the entire network are determined. During peak load periods, the predicted wind power of each wind resource area is corrected using the lower limit of the bandwidth; during off-peak load periods, the predicted wind power of each wind resource area is corrected using the upper limit of the bandwidth; and at other times, the predicted wind power of each wind resource area is corrected according to the average prediction error, thus obtaining the wind power of each wind resource area at the future time.

6. A wind power determination device based on probabilistic prediction, characterized in that, include: The search module is used to search the historical power prediction database of different wind resource areas for historical power prediction conditions that are similar to the power prediction conditions of the wind resource area at future times. The first determining module is used to determine the probability distribution of power prediction error for each wind resource area based on the wind power prediction error under historical power prediction conditions similar to the future power prediction conditions of the wind resource area. The second determining module is used to fit the probability distribution of power prediction error of the entire network by using Monte Carlo sampling method based on the probability distribution of power prediction error of each wind resource area. The correction module is used to correct the predicted wind power of each wind resource area at the future time according to the probability distribution of the power prediction error of the whole network, so as to obtain the wind power of each wind resource area at the future time. The search module includes: The acquisition unit is used to acquire the values ​​of the influence factors for each historical power prediction condition in the historical power prediction database of different wind resource areas. The determining unit is configured to confirm that the historical power prediction condition is similar to the future power prediction condition if the value of the influence factor of the historical power prediction condition and the value of the influence factor of the future power prediction condition meet the preset constraint conditions. The influencing factors include the ratio of the predicted power at the prediction time to the wind power installed capacity of the wind resource area at that prediction time, the ratio of the predicted power at the adjacent time before the prediction time to the wind power installed capacity of the wind resource area at that adjacent time, and the ratio of the predicted power at the adjacent time after the prediction time to the wind power installed capacity of the wind resource area at that adjacent time.

7. The apparatus according to claim 6, characterized in that, The device further includes: The third determining module is used to determine the day-ahead power generation plan of the power grid based on the wind power output of each wind resource area at the future time.

8. The apparatus according to claim 6, characterized in that, The influencing factors include at least two types; the determining unit is specifically used for: If the values ​​of each influencing factor under the historical power prediction conditions meet the corresponding preset constraints with the values ​​of the corresponding influencing factors under the power prediction conditions at future times, then the historical power prediction conditions are confirmed to be similar to the current power prediction conditions.

9. The apparatus according to claim 6, characterized in that, The correction module is specifically used for: Based on the probability distribution of the power prediction error of the entire network, the average prediction error, upper limit of bandwidth, and lower limit of bandwidth of the wind power of the entire network are determined. During peak load periods, the predicted wind power of each wind resource area is corrected using the lower limit of the bandwidth; during off-peak load periods, the predicted wind power of each wind resource area is corrected using the upper limit of the bandwidth; and at other times, the predicted wind power of each wind resource area is corrected according to the average prediction error, thus obtaining the wind power of each wind resource area at the future time.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images