A new energy output prediction method, device, equipment and storage medium

Abstract

The application discloses a new energy output prediction method, device and equipment and a storage medium, and is applied to the field of new energy output prediction. The method obtains predicted wind speed data and predicted irradiance data of a daily preset time interval, obtains a wind power output value prediction value of a corresponding time point through segmented fitting, obtains a photovoltaic output prediction value of the corresponding time point through fitting, multiplies the obtained daily total photovoltaic output prediction value by the proportion of the photovoltaic output prediction value of each time point in the sum of the photovoltaic output prediction values of all time points of the day, and obtains a target photovoltaic output prediction value. The application calculates the wind power output prediction value of each time point through segmented fitting, corrects the photovoltaic output prediction value of each time point by using the daily total photovoltaic output prediction value, and avoids the problems that the correction of the daily total photovoltaic output prediction result is not considered in the prior art, and the prediction result is inaccurate due to different wind speed intervals and different influences of wind speed on wind power output.

Description

Technical Field

[0001] This invention relates to the field of new energy power output prediction, and in particular to a new energy power output prediction method, apparatus, equipment and computer-readable storage medium. Background Technology

[0002] With my country's economic development, its energy structure is gradually changing, with the proportion of new energy sources showing an increasing trend. However, the output of new energy sources is greatly affected by meteorological factors and is unstable. Accurate forecasting of new energy output helps to rationally plan electricity consumption and the output of thermal power units, improve grid operating efficiency, and reduce operating costs. Therefore, accurate forecasting methods for new energy output are particularly important.

[0003] Currently, the main method for predicting renewable energy output is to input meteorological data such as wind speed and irradiance into machine learning or regression models. The main approach involves inputting historical actual irradiance, wind speed, and other meteorological data, along with renewable energy output data, into the model. The model parameters are then calculated to obtain the prediction model, and finally, meteorological forecast data is input into the prediction model to obtain the predicted renewable energy output value. However, current regression model prediction methods fail to consider corrections for the predicted daily total photovoltaic output, leading to overall overestimation or underestimation of photovoltaic output over a given period. Furthermore, they fail to account for the varying impacts of wind speed on wind power output across different time zones, resulting in a relatively coarse model. Summary of the Invention

[0004] The purpose of this invention is to provide a method, apparatus, device, and readable storage medium for predicting renewable energy output, which is applied in the field of renewable energy output prediction. This method predicts renewable energy output by correcting the total daily forecast value of photovoltaic output and performing piecewise linear regression fitting on the wind power output forecast value. This avoids the problems of inaccurate renewable energy output prediction caused by the failure to consider the correction of the prediction deviation of the total daily photovoltaic output value and the different effects of wind speed on wind power output in different intervals in the prior art.

[0005] To achieve the above objectives, the present invention provides a method for predicting the output of new energy sources, comprising:

[0006] Obtain daily predicted wind speed data and predicted irradiance data extracted at preset time intervals;

[0007] The predicted wind speed data is input into a piecewise function obtained by fitting the historical wind speed data and historical wind power output values ​​at the corresponding time point to obtain the predicted wind power output value at the corresponding time point.

[0008] The predicted irradiance data is input into a first function that is fitted by the historical irradiance data and historical photovoltaic output value at the corresponding time point to obtain the predicted photovoltaic output value at the corresponding time point of the day.

[0009] Obtain the predicted total daily photovoltaic output value, and multiply the predicted total daily photovoltaic output value by the proportion of the predicted photovoltaic output value at each time point to the sum of the predicted photovoltaic output values ​​at all time points on that day, to obtain the target predicted photovoltaic output value;

[0010] The predicted wind power output and the predicted target photovoltaic output are added together at each time point to obtain the predicted new energy output at each time point.

[0011] Optionally, the step of inputting the predicted wind speed data into a piecewise function obtained by piecewise fitting of historical wind speed data and historical wind power output values ​​at the corresponding time point to obtain the predicted wind power output value for the current time point includes:

[0012] The historical wind speed data and historical wind power output values ​​at the corresponding time points are subjected to three-segment linear fitting to obtain a three-segment function;

[0013] The predicted wind speed data is input into the three-segment function to obtain the predicted wind power output value for the corresponding time point of the day.

[0014] Optionally, obtaining the predicted total daily photovoltaic output includes:

[0015] Obtain the predicted daily total photovoltaic output residual and the annual trend value of daily total photovoltaic output, and then sum the predicted daily total photovoltaic output residual and the annual trend value of daily total photovoltaic output algebraically to obtain the predicted daily total photovoltaic output value.

[0016] Optionally, obtaining the residual of the predicted daily total photovoltaic output includes:

[0017] Obtain the predicted daily total irradiance residual, and input the predicted daily total irradiance residual into a second function that is linearly fitted by the historical daily total irradiance residual and the historical daily total photovoltaic output residual to obtain the predicted daily total photovoltaic output residual;

[0018] The historical daily total photovoltaic output residual is calculated by subtracting the annual trend value of the daily total photovoltaic output on the current day from the historical daily total irradiance.

[0019] Optionally, obtaining the annual trend value of the total daily photovoltaic output includes:

[0020] Input the date index into the third function, which is a univariate quadratic fit between the historical daily total photovoltaic output value and the historical date index, to obtain the annual trend value of the daily total photovoltaic output for that day.

[0021] Optionally, obtaining the predicted daily total irradiance residual includes:

[0022] Input the date index into the fourth function, which is a univariate quadratic fit between the historical daily total irradiance and the historical date index, to obtain the annual trend value of the daily total irradiance.

[0023] The predicted total daily irradiance residual is obtained by subtracting the annual trend value of the total daily irradiance from the predicted total daily irradiance.

[0024] Optionally, obtaining daily predicted wind speed data and predicted irradiance data extracted at preset time intervals includes:

[0025] Obtain the daily predicted wind speed data and predicted irradiance data extracted at 15-minute intervals.

[0026] To achieve the aforementioned objective, the present invention also provides a new energy output prediction device, comprising:

[0027] The data extraction module is used to acquire daily predicted wind speed data and predicted irradiance data extracted at preset time intervals;

[0028] The wind power calculation module is used to input the predicted wind speed data into a piecewise function obtained by piecewise fitting of historical wind speed data and historical wind power output values ​​at the corresponding time point, so as to obtain the predicted wind power output value at the corresponding time point of the day.

[0029] The photovoltaic calculation module is used to input the predicted irradiance data into a first function obtained by fitting the historical irradiance data and historical photovoltaic output values ​​at the corresponding time point, so as to obtain the predicted photovoltaic output value at the corresponding time point of the day;

[0030] The photovoltaic correction module is used to obtain the predicted value of the total daily photovoltaic output, and multiply the predicted value of the total daily photovoltaic output by the proportion of the predicted value of the photovoltaic output at each time point to the sum of the predicted values ​​of the photovoltaic output at all time points on that day, to obtain the target predicted value of the photovoltaic output.

[0031] The new energy calculation module is used to add the predicted wind power output value and the predicted target photovoltaic power output value at each time point to obtain the predicted new energy output value at each time point.

[0032] To achieve the above objectives, the present invention also provides a new energy output prediction device, comprising:

[0033] Memory, used to store computer programs;

[0034] A processor is used to implement any of the above-described new energy output prediction methods when executing the computer program.

[0035] To achieve the above objectives, the present invention also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement any of the above-described new energy output prediction methods.

[0036] As can be seen, the present invention calculates the wind power output prediction value at each time point by segmented fitting, and corrects the photovoltaic output prediction value at each time point using the daily total photovoltaic output prediction value. This avoids the problems of inaccurate prediction results caused by the failure to consider the correction of the daily total photovoltaic output prediction result and the different effects of wind speed on wind power output in different intervals in the prior art. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention 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 embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0038] Figure 1 A flowchart of a new energy output prediction method provided in an embodiment of the present invention;

[0039] Figure 2 A comparison chart of new energy output prediction provided in an embodiment of the present invention;

[0040] Figure 3 A comparison chart of the average absolute error of new energy output prediction provided in an embodiment of the present invention;

[0041] Figure 4 A specific embodiment of a new energy output prediction method provided by the present invention is shown in the figure.

[0042] Figure 5 This is a structural block diagram of a new energy output prediction device provided in an embodiment of the present invention. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] Based on extensive research, the widely used linear regression model for predicting renewable energy output exhibits poor accuracy in wind power output prediction when wind speeds are varying, and often shows an overall overestimation or underestimation of photovoltaic (PV) output. To address this, this invention aims to improve existing linear regression model prediction methods by comprehensively considering the impact of wind speeds across different ranges on wind power output and correcting the daily total PV output prediction results. This will correct the overall fluctuations in the PV output prediction curve, ultimately improving the accuracy of renewable energy output prediction.

[0045] The following combination Figure 1 , Figure 1 A flowchart of a new energy output prediction method provided in an embodiment of the present invention, the method may include:

[0046] S101: Obtain daily predicted wind speed data and predicted irradiance data extracted at preset time intervals.

[0047] This embodiment does not limit the size of the preset time interval for data extraction. Users can set it themselves based on data density and actual usage scenarios. Generally, a time interval that divides the 1440 minutes of a day evenly can be selected, such as 10 minutes, 15 minutes, or 20 minutes. Since the typical time interval for collecting wind power output data and photovoltaic output data is 15 minutes, the preset time interval in this application can be 15 minutes. Based on the preset time interval, the day can be divided into multiple time points. For example, with a preset time interval of 15 minutes, the day can be divided into 96 time points, with 0:15 as the first time point and 0:30 as the second time point. This embodiment can extract the predicted wind speed data and predicted irradiance data corresponding to each time point of the day. Furthermore, to make the obtained new energy output prediction values ​​more accurate, this embodiment can also smooth the predicted wind speed data and predicted irradiance data before or after extraction and remove some invalid data points.

[0048] S102: Input the predicted wind speed data into a piecewise function obtained by piecewise fitting of the historical wind speed data and historical wind power output values ​​at the corresponding time point, and obtain the predicted wind power output value for the corresponding time point.

[0049] Because wind speed has varying impacts on wind power output across different time zones, this embodiment aims to improve the accuracy of wind power output predictions at each time point. The predicted wind speed data is input into a piecewise function obtained by fitting historical wind speed data and historical wind power output values ​​at the corresponding time point. The historical wind speed data and historical wind power output values ​​used in this embodiment are the historical wind speed data and historical wind power output values ​​at the corresponding time points. For example, if the predicted wind speed data is the predicted wind speed data at 0:15 on the current day, then the piecewise fitted historical wind speed data and historical wind power output values ​​are the actual wind speed data and actual wind power output values ​​collected at 0:15 every day in the historical time period. The days are arranged by date index; for example, the date index of day 1 in 2022 is 1, the date index of day 2 is 2, and the date index of day 365 is 365. At this point, by inputting the predicted wind speed data at 0:15 on the day with date index 1 into the piecewise fitting function corresponding to that time point, the predicted wind power output value at 0:15 on that day can be obtained. Similarly, by inputting the predicted wind speed data at 0:30 on the day with date index 2 into the piecewise fitting function corresponding to that time point, the predicted wind power output value at 0:30 on that day can be obtained.

[0050] Furthermore, this embodiment does not limit the method of piecewise fitting, nor does it limit the number of segment intervals. Generally, the least squares method can be used to perform piecewise linear fitting on historical wind power data and historical wind power output values. The segment intervals can be divided into 3, and the piecewise function can be:

[0051]

[0052] Among them, v i For each time point, the predicted wind speed data is provided. When i is 1, it is the first time point; when i is 2, it is the second time point. x1 and x2 are the boundary values ​​for dividing the wind speed intervals. This embodiment does not limit the setting method of these boundary values ​​and can be set by the user. windiThis represents the predicted wind speed value at the corresponding time point. When the input is the predicted wind speed data corresponding to the i-th time point, it is fitted piecewise from the historical irradiance data of the i-th time point in the corresponding historical time period and its corresponding historical wind power output value. For example, when i is 1, k1,b1 can be obtained by fitting the historical wind speed data less than or equal to x1 corresponding to the first time point in the historical time period and its corresponding wind power output value; k2,b2 can be obtained by fitting the historical wind speed data greater than x1 and less than x2 corresponding to the first time point in the historical time period and its corresponding wind power output value; k3,b3 can be obtained by fitting the historical wind speed data greater than x1 and less than x2 corresponding to the first time point in the historical time period and its corresponding wind power output value; and k3,b3 can be obtained by fitting the historical wind speed data of the first time point in the historical time period and its corresponding wind power output value. The values ​​are obtained by fitting historical wind speed data greater than or equal to x2 with their corresponding wind power output values. When i is 2, k1,b1 can be obtained by fitting historical wind speed data less than or equal to x1 corresponding to the second time point in the historical time period with their corresponding wind power output values; k2,b2 can be obtained by fitting historical wind speed data greater than x1 and less than x2 corresponding to the second time point in the historical time period with their corresponding wind power output values; and k3,b3 can be obtained by fitting historical wind speed data greater than or equal to x2 corresponding to the second time point in the historical time period with their corresponding wind power output values. Here, k1,b1, k2,b2, and k3,b3 can be the slope and intercept of the function.

[0053] Furthermore, the interval boundary value corresponding to each time point can be set based on the historical data corresponding to that time point; that is, the interval boundary value for each time point can be the same or different. When the predicted wind speed data of the first time point with the date index 1 of the year to be predicted is input into the piecewise function corresponding to the first time point, the predicted wind power output value of the first time point with date index 1 can be obtained; when the predicted wind speed data of the first time point with the date index 2 of the year to be predicted is input into the piecewise function corresponding to the second time point, the predicted wind power output value of the first time point with date index 2 can be obtained.

[0054] S103: Input the predicted irradiance data into the first function obtained by fitting the historical irradiance data and historical photovoltaic output values ​​at the corresponding time point to obtain the predicted photovoltaic output value for the corresponding time point.

[0055] This embodiment obtains the predicted photovoltaic output value for the corresponding time point by inputting the predicted irradiance data into a first function that is fitted with historical irradiance data and historical photovoltaic output values ​​at the corresponding time point. In this embodiment, the historical irradiance data and historical photovoltaic output values ​​used are the historical irradiance data and historical photovoltaic output values ​​at the corresponding time point. For example, if the predicted irradiance data is the predicted irradiance data at 0:15 on the current day, then the historical irradiance data and historical photovoltaic output values ​​used for fitting are the actual irradiance data and actual photovoltaic output values ​​collected at 0:15 every day in the historical time period. This embodiment does not limit the fitting method; generally, the least squares method can be used to perform piecewise linear fitting of the historical irradiance data and historical photovoltaic output values. The first function can then be:

[0056] P radi =k*R i +b;

[0057] Among them, R i The predicted irradiance data for each time point, P radi The input is the predicted photovoltaic output value at the corresponding time point. When the input is the predicted irradiance data corresponding to the i-th time point, k and b can be the slope and intercept of the function. Then k and b can be obtained by fitting the historical irradiance data of the i-th time point in the corresponding historical time period with its corresponding historical photovoltaic output value.

[0058] The predicted irradiance data is input into a first function that is fitted with historical irradiance data and historical photovoltaic output values ​​at the corresponding time point to obtain the predicted photovoltaic output value for that time point. For example, when the predicted irradiance data of the first time point with the date index 1 of the year to be predicted is input into the first function corresponding to the first time point, the predicted photovoltaic output value of the first time point with the date index 1 can be obtained; when the predicted irradiance data of the first time point with the date index 2 of the year to be predicted is input into the first function corresponding to the first time point, the predicted photovoltaic output value of the first time point with the date index 2 can be obtained.

[0059] S104: Obtain the predicted total daily photovoltaic output value, multiply the predicted total daily photovoltaic output value by the proportion of the predicted photovoltaic output value at each time point to the sum of the predicted photovoltaic output values ​​at all time points on that day, and obtain the target photovoltaic output predicted value.

[0060] This embodiment obtains the predicted daily total photovoltaic (PV) output value and uses it to correct the PV output prediction value at each time point, making the target PV output prediction value more accurate. This embodiment does not limit the method of obtaining the predicted daily total PV output value. Generally, it can be obtained by obtaining the predicted daily total PV output residual and the annual trend value of the daily total PV output, and then algebraically summing and adding these two values ​​to obtain the predicted daily total PV output value, as shown in the following formula:

[0061] P rad-day =D power-rad +P rad-day-trend ;

[0062] Among them, P rad-day D represents the predicted total daily photovoltaic power output. power-rad To predict the residual of total daily photovoltaic output, p rad-day-trend This represents the annual trend of Japan's total photovoltaic power output.

[0063] This embodiment does not limit the calculation method of the predicted daily total photovoltaic output residual. Generally, it can be obtained by obtaining the predicted daily total irradiance residual and inputting it into a second function that is linearly fitted by the historical daily total irradiance residual and the historical daily total photovoltaic output residual. The predicted daily total photovoltaic output residual can be obtained by:

[0064] D power-rad =k4*D rad +b4;

[0065] Among them, D power-rad To predict the residual of total daily photovoltaic output, D rad To predict the daily total irradiance residual, k4,b4 can be the slope and intercept of the function. k4,b4 is obtained by linear fitting of the historical daily total irradiance residual and the historical daily total photovoltaic output residual. By inputting the predicted daily total irradiance residual into the second function, the corresponding predicted daily total photovoltaic output residual can be obtained. For example, by inputting the predicted irradiance residual with date index 1 into the second function, the predicted daily total photovoltaic output residual for that day can be obtained.

[0066] Furthermore, this embodiment does not limit the calculation method of the predicted daily total irradiance residual. Generally, the annual trend value of the daily total irradiance can be obtained by inputting the date index into a fourth function that performs a univariate quadratic fitting of the historical daily total irradiance and the historical date index. Subtracting the annual trend value of the daily total irradiance from the predicted daily total irradiance yields the predicted daily total irradiance residual. The fourth function can be:

[0067] R day-trend =a1x 2 +b1x+c1;

[0068] Among them, R day-trend This function represents the annual trend of total daily irradiance, where x is the date index, and a1, b1, and c1 are the parameter values ​​of the function. a1, b1, and c1 are obtained by performing a univariate quadratic fitting of historical total daily irradiance and historical date indices. When a date index is input to this fourth function, the function outputs the annual trend of total daily irradiance for the corresponding date. For example, when x is 1, the function outputs the annual trend of total daily irradiance for the first day of that year.

[0069] Subtracting the annual trend value of the total daily irradiance from the predicted total daily irradiance yields the residual of the predicted total daily irradiance, as shown below:

[0070] D rad =R day -R day-trend ;

[0071] Among them, D rad To predict the residual of total daily irradiance, R day-trend R represents the annual trend value of total daily irradiance. day To predict the total daily irradiance, the obtained residual of the predicted total daily irradiance is input into the second function to obtain the corresponding residual of the predicted total daily photovoltaic output. Furthermore, in this formula, if R... day Given the actual historical total daily irradiance, the obtained D rad This transformation yields the actual historical daily total irradiance residual, which can be used to fit the second function.

[0072] This embodiment does not limit the calculation method of the annual trend value of the total daily photovoltaic output. Generally, the date index can be input into a third function that performs a univariate quadratic fitting of the historical total daily photovoltaic output value and the historical date index to obtain the annual trend value of the total daily photovoltaic output for the current day. This third function can be:

[0073] P rad-day-trend =a2x 2 +b²x+c²;

[0074] Among them, P rad-day-trend This function represents the annual trend of total daily photovoltaic output, where x is the date index, and a2, b2, and c2 are the parameters of the function. a2, b2, and c2 are obtained by performing a univariate quadratic fitting on the historical total daily photovoltaic output and the historical date index. When the date index is input to this third function, the function outputs the annual trend of total daily photovoltaic output for the corresponding date. For example, when x is 1, the function outputs the annual trend of total daily photovoltaic output for the first day of that year.

[0075] Furthermore, the actual historical total daily photovoltaic output residuals used to fit the second function can be calculated using the annual trend value of the total daily photovoltaic output. The historical total daily photovoltaic output residuals can be obtained by subtracting the annual trend value of the total daily photovoltaic output from the historical total daily photovoltaic output value, as shown below:

[0076] D real-power-rad =P real-rad-day -P rad-day-trend ;

[0077] Among them, D real-power-rad For the historical daily total photovoltaic power output residual, P real-rad-day P represents the historical daily total photovoltaic power output. rad-day-trend This represents the annual trend of Japan's total photovoltaic power output.

[0078] Finally, the algebraic sums of the calculated residuals of the predicted daily total photovoltaic output and the annual trend value of the daily total photovoltaic output are added together to obtain the predicted value of the daily total photovoltaic output. The predicted value of the daily total photovoltaic output is then multiplied by the proportion of the predicted photovoltaic output at each time point of the day to the sum of the predicted photovoltaic output at all time points of the day, as shown below:

[0079]

[0080] Among them, P new-radi P represents the predicted target photovoltaic power output at the corresponding time point. rad-day P represents the predicted total daily photovoltaic output. radi The summation formula is the sum of the predicted photovoltaic output values ​​at the corresponding time points, where m is the total number of time points on that day.

[0081] S105: Add the predicted wind power output and the predicted target photovoltaic power output at each time point to obtain the predicted new energy output at each time point.

[0082] By adding the predicted wind power output at each time point to the target predicted photovoltaic output, the predicted renewable energy output at each time point can be obtained. The predicted renewable energy output at each time point obtained using the method of this invention is more accurate than the traditional linear regression model prediction method, which neither uses piecewise fitting to calculate the predicted wind power output at each time point nor uses the predicted daily total photovoltaic output to correct the predicted photovoltaic output at each time point. Specifically, it can be seen as follows... Figure 2 As shown, Figure 2 The data in this article comes from the forecast results of new energy power output in Shanxi Province, and has a smaller absolute error. For details, see below. Figure 3 As shown, Figure 3 The data in this article comes from the daily average error of the forecast results for new energy power output in Shanxi Province.

[0083] This embodiment calculates the predicted wind power output at each time point by segmented fitting, and corrects the predicted photovoltaic output at each time point using the predicted total daily photovoltaic output. This avoids the problems in the prior art where the correction of the predicted total daily photovoltaic output is not considered and the prediction results are inaccurate due to the different effects of wind speed on wind power output in different intervals.

[0084] The following combination Figure 4 , Figure 4 This is a specific embodiment of a new energy output prediction method provided by the present invention. The predicted new energy output value can be date index 6, the preset time interval can be 15 minutes, and the time point can be the first time point. This specific embodiment may include:

[0085] 1. Obtain daily predicted wind speed data and predicted irradiance data extracted at 15-minute intervals.

[0086] 2. Input the predicted wind speed data at 0:15 on the 6th day of the predicted year into a piecewise function fitted by the historical wind speed data and historical wind power output values ​​at 0:15 on the historical days, and obtain the predicted wind power output value at 0:15 on the 6th day of the year to be predicted.

[0087] 3. Input the predicted irradiance data at 0:15 on the 6th day of the predicted year into the first function that is fitted by the historical irradiance data at 0:15 on the historical days and the historical photovoltaic output value, and obtain the predicted photovoltaic output value at 0:15 on the 6th day of the year to be predicted.

[0088] 4. Obtain the predicted total daily photovoltaic output value, multiply the predicted total daily photovoltaic output value by the sum of the predicted photovoltaic output value at this time point and the predicted photovoltaic output value at all time points on the 6th day, to obtain the predicted target photovoltaic output value at this time point.

[0089] 5. Add the predicted wind power output at this time point to the predicted target photovoltaic power output to obtain the predicted new energy output value.

[0090] The following describes the new energy output prediction device, equipment, and storage medium provided by the embodiments of the present invention. The new energy output prediction device, equipment, and storage medium described below can be referred to in correspondence with the new energy output prediction method described above.

[0091] The following combination Figure 5 , Figure 5 This invention provides a structural block diagram of a new energy output prediction device, which may include:

[0092] The data extraction module 100 is used to acquire daily predicted wind speed data and predicted irradiance data extracted at preset time intervals;

[0093] The wind power calculation module 200 is used to input the predicted wind speed data into a piecewise function obtained by piecewise fitting of the historical wind speed data and historical wind power output values ​​at the corresponding time point, so as to obtain the predicted wind power output value at the corresponding time point of the day.

[0094] The photovoltaic calculation module 300 is used to input the predicted irradiance data into a first function obtained by fitting the historical irradiance data and historical photovoltaic output value at the corresponding time point, so as to obtain the predicted photovoltaic output value at the corresponding time point of the day.

[0095] The photovoltaic correction module 400 is used to obtain the daily total photovoltaic output forecast value, and multiply the daily total photovoltaic output forecast value by the proportion of the photovoltaic output forecast value at each time point to the sum of the photovoltaic output forecast values ​​at all time points on that day, to obtain the target photovoltaic output forecast value.

[0096] The new energy calculation module 500 is used to add the predicted wind power output and the predicted target photovoltaic output at each time point to obtain the predicted new energy output at each time point.

[0097] Based on the above embodiments, the present invention calculates the predicted wind power output at each time point by segmented fitting, and corrects the predicted photovoltaic output at each time point using the predicted total daily photovoltaic output. This avoids the problems in the prior art where the correction of the predicted total daily photovoltaic output is not considered and the prediction results are inaccurate due to the different effects of wind speed on wind power output in different intervals.

[0098] Based on the above embodiments, the wind power calculation module 200 may include:

[0099] The three-segment fitting unit is used to perform three-segment linear fitting on the historical wind speed data and historical wind power output values ​​at the corresponding time points to obtain a three-segment function.

[0100] The three-segment calculation unit is used to input the predicted wind speed data into the three-segment function to obtain the predicted wind power output value at the corresponding time point of the day.

[0101] Based on the above embodiments, the photovoltaic correction module 400 may include:

[0102] The daily total photovoltaic output prediction submodule is used to obtain the predicted daily total photovoltaic output residual and the annual trend value of daily total photovoltaic output. The predicted daily total photovoltaic output residual and the annual trend value of daily total photovoltaic output are algebraically summed and added together to obtain the predicted daily total photovoltaic output value.

[0103] Based on the above embodiments, the daily total photovoltaic output prediction submodule includes:

[0104] The daily total photovoltaic output residual unit is used to obtain the predicted daily total irradiance residual. The predicted daily total irradiance residual is input into the second function that is linearly fitted by the historical daily total irradiance residual and the historical daily total photovoltaic output residual to obtain the predicted daily total photovoltaic output residual.

[0105] The historical daily total photovoltaic output residual is calculated by subtracting the annual trend value of the search day's total photovoltaic output from the historical daily total irradiance.

[0106] Based on the above embodiments, the daily total photovoltaic output prediction submodule includes:

[0107] The daily total photovoltaic output annual trend value unit is used to input the date index into the third function, which is a univariate quadratic fitting of the historical daily total photovoltaic output value and the historical date index, to obtain the daily total photovoltaic output annual trend value for the current day.

[0108] Based on the above embodiments, the daily total photovoltaic output residual unit includes:

[0109] The sub-unit for annual variation trend of total daily irradiance is used to input the date index into the fourth function, which is fitted by a univariate quadratic fit of historical total daily irradiance and historical date index, to obtain the annual variation trend value of total daily irradiance for the current day.

[0110] The predicted total daily irradiance residual sub-unit is used to subtract the annual trend value of the total daily irradiance from the predicted total daily irradiance to obtain the predicted total daily irradiance residual.

[0111] Based on the above embodiments, the data extraction module 100 includes:

[0112] The data extraction unit is used to acquire daily predicted wind speed data and predicted irradiance data extracted at 15-minute intervals.

[0113] Based on the above embodiments, the present invention also provides a new energy output prediction device. This device may include a memory and a processor. The memory stores a computer program, and when the processor calls the computer program in the memory, it can implement the steps provided in the above embodiments. Of course, the device may also include various necessary network interfaces, power supplies, and other components.

[0114] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by an execution terminal or processor, it can implement the new energy output prediction method provided in the embodiments of the present invention. The storage medium may include various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0115] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0116] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0117] The present invention provides a detailed description of a new energy output prediction method, apparatus, device, and storage medium. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, those skilled in the art will recognize that there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A new energy output prediction method, characterized in that, include: Obtain daily predicted wind speed data and predicted irradiance data extracted at preset time intervals; The predicted wind speed data is input into a piecewise function obtained by fitting historical wind speed data and historical wind power output values ​​at the corresponding time point to obtain a predicted wind power output value for the current time point; this includes: performing a three-segment linear fit on the historical wind speed data and historical wind power output values ​​at the corresponding time point to obtain a three-segment function; and inputting the predicted wind speed data into the three-segment function to obtain a predicted wind power output value for the current time point. The predicted irradiance data is input into a first function that is fitted by the historical irradiance data and historical photovoltaic output value at the corresponding time point to obtain the predicted photovoltaic output value at the corresponding time point of the day. The process involves obtaining a predicted daily total photovoltaic (PV) output value, multiplying this predicted value by the proportion of the predicted PV output value at each time point to the sum of the predicted PV output values ​​at all time points on that day, to obtain a target PV output prediction value. This includes: obtaining the predicted daily total PV output residual and the annual trend value of the daily total PV output; summing these two values ​​algebraically to obtain the predicted daily total PV output value; and obtaining the predicted daily total PV output residual, which includes: obtaining the predicted daily total irradiance residual; inputting this residual into a second function that is linearly fitted by historical daily total irradiance residuals and historical daily total PV output residuals to obtain the predicted value. The residual of total daily photovoltaic output; wherein, the residual of total daily photovoltaic output is calculated by subtracting the annual trend value of the total daily photovoltaic output of the current day from the historical total daily irradiance; obtaining the annual trend value of the total daily photovoltaic output includes: inputting the date index into a third function that performs a univariate quadratic fitting of the historical total daily photovoltaic output value and the historical date index to obtain the annual trend value of the total daily photovoltaic output of the current day; obtaining the residual of predicted total daily irradiance includes: inputting the date index into a fourth function that performs a univariate quadratic fitting of the historical total daily irradiance and the historical date index to obtain the annual trend value of the total daily irradiance of the current day; subtracting the annual trend value of the total daily irradiance from the predicted total daily irradiance to obtain the residual of predicted total daily irradiance; The predicted wind power output and the predicted target photovoltaic output are added together at each time point to obtain the predicted new energy output at each time point.

2. The new energy output prediction method according to claim 1, characterized in that, The acquisition of daily predicted wind speed data and predicted irradiance data extracted at preset time intervals includes: Obtain the daily predicted wind speed data and predicted irradiance data extracted at 15-minute intervals.

3. A new energy output prediction device, characterized in that, include: The data extraction module is used to acquire daily predicted wind speed data and predicted irradiance data extracted at preset time intervals; The wind power calculation module is used to input the predicted wind speed data into a piecewise function obtained by piecewise fitting of historical wind speed data and historical wind power output values ​​at the corresponding time point, to obtain the predicted wind power output value for the current time point; including: performing a three-segment linear fitting of the historical wind speed data and historical wind power output values ​​at the corresponding time point to obtain a three-segment function; inputting the predicted wind speed data into the three-segment function to obtain the predicted wind power output value for the current time point; The photovoltaic calculation module is used to input the predicted irradiance data into a first function obtained by fitting the historical irradiance data and historical photovoltaic output values ​​at the corresponding time point, so as to obtain the predicted photovoltaic output value at the corresponding time point of the day; The photovoltaic correction module is used to obtain the predicted daily total photovoltaic output value, and multiply the predicted daily total photovoltaic output value by the proportion of the predicted photovoltaic output value at each time point to the sum of the predicted photovoltaic output values ​​at all time points on that day, to obtain the target predicted photovoltaic output value; including: obtaining the predicted daily total photovoltaic output residual and the annual variation trend value of the daily total photovoltaic output, and adding the predicted daily total photovoltaic output residual and the annual variation trend value of the daily total photovoltaic output algebraically to obtain the predicted daily total photovoltaic output value; obtaining the predicted daily total photovoltaic output residual includes: obtaining the predicted daily total irradiance residual, and inputting the predicted daily total irradiance residual into a second function that is linearly fitted by the historical daily total irradiance residual and the historical daily total photovoltaic output residual, to obtain The method involves obtaining the predicted daily total photovoltaic output residual; wherein the historical daily total photovoltaic output residual is calculated by subtracting the annual trend value of the daily total photovoltaic output for the current day from the historical daily total irradiance; obtaining the annual trend value of the daily total photovoltaic output includes: inputting the date index into a third function that performs a univariate quadratic fitting between the historical daily total photovoltaic output value and the historical date index to obtain the annual trend value of the daily total photovoltaic output for the current day; obtaining the predicted daily total irradiance residual includes: inputting the date index into a fourth function that performs a univariate quadratic fitting between the historical daily total irradiance and the historical date index to obtain the annual trend value of the daily total irradiance for the current day; and subtracting the annual trend value of the daily total irradiance from the predicted daily total irradiance to obtain the predicted daily total irradiance residual; The new energy calculation module is used to add the predicted wind power output value and the predicted target photovoltaic power output value at each time point to obtain the predicted new energy output value at each time point.

4. A new energy output prediction device, characterized in that, include: Memory, used to store computer programs; A processor is configured to implement the new energy output prediction method as described in any one of claims 1 or 2 when executing the computer program.

5. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, implement the new energy output prediction method as described in any one of claims 1 or 2.

Images