A comprehensive evaluation method and device for power prediction deviation of a photovoltaic power station

Abstract

In the comprehensive evaluation method and device for power prediction deviation of a photovoltaic power station and the storage medium, actual power data at each time in a daily average sunshine time interval in a first preset time period of the photovoltaic power station is obtained, a daily average target time interval for calculating the power prediction deviation of the photovoltaic power station is obtained, actual power data and power prediction data at each time in the daily average target time interval in a second preset time period of the photovoltaic power station are obtained, a deviation index of the power prediction of the photovoltaic power station is calculated according to the power prediction data and the actual power data at each time in the daily average target time interval in the second preset time period, and the deviation index is used to comprehensively evaluate the power prediction deviation of the photovoltaic power station. In the application, the result of the comprehensive evaluation of the power prediction deviation of the photovoltaic power station is more accurate, and the result of the comprehensive evaluation can be used to compare the advantages and disadvantages of different prediction results.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of photovoltaic power generation, and particularly relates to a comprehensive evaluation method and device for photovoltaic power station power prediction deviation and a storage medium. BACKGROUND

[0002] The photovoltaic power station needs to make power prediction in advance to guide the dispatching operation of the power system. However, due to reasons such as cloud cover, photovoltaic power generation has the characteristics of randomness, and the prediction is difficult, and the results of various photovoltaic power prediction systems are not the same. Therefore, the deviation of the photovoltaic power prediction needs to be calculated to objectively evaluate the accuracy of the photovoltaic power prediction according to the deviation of the photovoltaic power prediction, so as to promote the improvement of the accuracy of the photovoltaic power prediction.

[0003] At present, the existing deviation calculation method of photovoltaic power prediction mainly refers to the deviation calculation method of wind power prediction. According to the actual power and the predicted power in the past period of time, the deviation of power prediction is obtained by calculation. However, due to the influence of sunlight on photovoltaic power generation, there is actual power at the moment when the light is weak, but because the actual power is small, the power prediction may be directly zero. At this time, directly using the above method to calculate the deviation of the photovoltaic power prediction will cause the evaluation results to be generally deviated, and the advantages and disadvantages between different prediction results cannot be effectively compared. SUMMARY

[0004] The present application provides a comprehensive evaluation method and device for photovoltaic power station power prediction deviation and a storage medium to solve the technical problems in the above related technologies.

[0005] The first aspect embodiment of the present application provides a comprehensive evaluation method for photovoltaic power station power prediction deviation, comprising:

[0006] Obtaining the actual power data of each time point in the daily average sunshine time interval of the photovoltaic power station in the first preset time period to obtain the daily average target time interval for calculating the power prediction deviation of the photovoltaic power station;

[0007] Obtaining the actual power data and power prediction data of each time point in the daily average target time interval of the photovoltaic power station in the second preset time period;

[0008] According to the power prediction data and the actual power data of each time point in the daily average target time interval in the second preset time period, the deviation index of the photovoltaic power prediction is calculated;

[0009] The deviation index is used to comprehensively evaluate the power prediction deviation of the photovoltaic power station.

[0010] The second aspect embodiment of the present application provides a comprehensive evaluation device for photovoltaic power station power prediction deviation, comprising:

[0011] The first calculation module is configured to obtain, by using the actual power data at each time in the daily average sunshine time interval in the first preset time period of the photovoltaic power station, a daily average target time interval for calculating the power prediction deviation of the photovoltaic power station;

[0012] The acquisition module is configured to acquire the actual power data and the power prediction data at each time in the daily average target time interval in the second preset time period of the photovoltaic power station;

[0013] The second calculation module is configured to calculate, according to the power prediction data and the actual power data at each time in the daily average target time interval in the second preset time period, a deviation index of the power prediction of the photovoltaic power station.

[0014] The evaluation module is configured to comprehensively evaluate the power prediction deviation of the photovoltaic power station by using the deviation index.

[0015] The computer storage medium provided in the third aspect of the present application, wherein the computer storage medium stores computer executable instructions; the computer executable instructions are executed by the processor, and the method described in the first aspect above can be implemented.

[0016] The computer device provided in the fourth aspect of the present application, wherein it includes a memory, a processor and a computer program stored in the memory and executable on the processor, and when the processor executes the program, the method described in the first aspect above can be implemented.

[0017] The technical scheme provided by the embodiments of the present application at least brings the following beneficial effects:

[0018] In the comprehensive evaluation method, device and storage medium for power prediction deviation of a photovoltaic power station provided by the application, the actual power data at each time in the daily average sunshine time interval in a first preset time period of the photovoltaic power station is obtained, and a daily average target time interval for calculating the power prediction deviation of the photovoltaic power station is obtained. The actual power data and the power prediction data at each time in the daily average target time interval in a second preset time period of the photovoltaic power station are obtained. According to the power prediction data and the actual power data at each time in the daily average target time interval in the second preset time period, the deviation index of the power prediction of the photovoltaic power station is calculated. The deviation index is used to comprehensively evaluate the power prediction deviation of the photovoltaic power station. In the application, the actual power data at each time in the daily average sunshine time interval in the first preset time period of the photovoltaic power station is obtained. The time interval with weak light in the daily average sunshine time interval is deleted to obtain the daily average target time interval for calculating the power prediction deviation of the photovoltaic power station. The actual power data and the power prediction data at each time in the daily average target time are used to calculate the deviation index of the photovoltaic power station. Therefore, the result of the comprehensive evaluation of the power prediction deviation of the photovoltaic power station is more accurate, and the results of the comprehensive evaluation can be used to compare the advantages and disadvantages of different prediction results.

[0019] Additional aspects and advantages of the application will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following and / or by practice of the application. BRIEF DESCRIPTION OF DRAWINGS

[0020] The above and / or additional aspects and advantages of the application will become apparent and be more readily understood through consideration of the following description, taken in conjunction with the accompanying drawings, in which:

[0021] Figure 1 A flowchart of a comprehensive evaluation method for power prediction deviation of a photovoltaic power station according to an embodiment of the application is provided.

[0022] Figure 2 A structural diagram of a comprehensive evaluation device for power prediction deviation of a photovoltaic power station according to an embodiment of the application is provided. DETAILED DESCRIPTION

[0023] The embodiments of the application are described in detail below, examples of which are shown in the accompanying drawings, in which the same or similar reference signs represent the same or similar elements or elements with the same or similar functions throughout. The embodiments described below by referring to the accompanying drawings are exemplary and are intended to explain the application, and cannot be understood as a limitation of the application.

[0024] The comprehensive evaluation method and device for power prediction deviation of a photovoltaic power station of the embodiments of the application are described below with reference to the accompanying drawings.

[0025] Embodiment one

[0026] Figure 1 A flowchart of a comprehensive evaluation method of power prediction deviation of a photovoltaic power station according to an embodiment of the present application is shown in FIG. 1, which can include the following steps: Figure 1

[0027] In step 101, the actual power data at each time in the daily average sunshine time interval in the first preset time period of the photovoltaic power station is obtained to obtain the daily average target time interval for calculating the power prediction deviation of the photovoltaic power station.

[0028] In an embodiment of the present application, the meteorological monitoring data of the photovoltaic power station in the past year can be obtained, and the daily average sunshine time interval of the photovoltaic power station can be obtained by analyzing the direct irradiance and scattered irradiance historical data.

[0029] In an embodiment of the present application, the daily average sunshine time interval of the photovoltaic power station obtained by the above analysis can have the following two cases:

[0030] Case one: the daily average sunshine time interval of the photovoltaic power station is the daily average sunshine time interval of the whole year (for example, [6:00, 18:00));

[0031] Case two: the daily average sunshine time interval of the photovoltaic power station is the daily average sunshine time interval of each season, and there is a corresponding daily average sunshine time interval for each season, and each 3 months of the year is a season.

[0032] In an embodiment of the present application, the photovoltaic power station is affected by sunlight, and there is actual power at a time when the light is weak, but the power prediction can be directly zero because the actual power is small. If the actual power data and the power prediction data at each time in the daily average sunshine time interval are directly used to calculate the deviation, the deviation at the time when the light is weak will be large, resulting in that the evaluation result obtained by calculation is generally deviated. Therefore, before calculating the deviation index of the power prediction of the photovoltaic power station, the time that affects the calculation accuracy of the deviation needs to be removed to obtain the daily average target time interval.

[0033] Specifically, in an embodiment of the present application, the method for obtaining the daily average target time interval for calculating the power prediction deviation of the photovoltaic power station from the actual power data at each time in the daily average sunshine time interval in the first preset time period of the photovoltaic power station can include the following steps:

[0034] In step 1011, the average actual power at each time in the third preset time period and the fourth preset time period in the daily average sunshine time interval is calculated according to the actual power data at each time in the daily average sunshine time interval in the first preset time period.

[0035] ​In one embodiment of the present application, the sunshine duration is related to seasons, and the sunshine duration corresponding to different seasons, based on which the first preset time period can be one year in the past so as to consider the sunshine duration of all seasons.

[0036] In one embodiment of the present application, the time instant affecting the deviation precision of the photovoltaic power station is the time instant when the sunshine intensity is low, the actual power data is low, and the corresponding power prediction data is zero. Through analysis, it can be concluded that the corresponding time instant is the time instant when the sunshine just starts and the time instant when the sunshine is about to disappear. Based on this, the third preset time period of the daily average sunshine duration interval can be two hours when the daily average sunshine duration starts, the fourth preset time period can be two hours when the daily average sunshine duration is about to end, and each 15 minutes can be a time instant, so that the third preset time period and the fourth preset time period each include 8 time instants.

[0037] For example, in one embodiment of the present application, assuming that the daily average sunshine duration interval is [6:00, 18:00], the third preset time period is [6:00, 8:00), and a time instant is 15 minutes, then the 8 time instants in the third preset time period are 6:00, 6:15, 6:30, 6:45, 7:00, 7:15, 7:30, 7:45, and the fourth preset time period is (16:00, 18:00], and the 8 time instants in the fourth preset time period are 16:15, 16:30, 16:45, 17:00, 17:15, 17:30, 17:45, and 18:00.

[0038] Further, in one embodiment of the present application, the average value of the actual power data corresponding to each time instant in the third preset time period in the past one year can be obtained, to obtain the average power value P i of the i-th time instant in the third preset time period, where i=1, 2, …, n, n is the total number of time instants in the third preset time period, and the average value of the actual power data corresponding to each time instant in the fourth preset time period in the past one year can be obtained, to obtain the average power value M j of the j-th time instant in the fourth preset time period, where j=1, 2, …, m, m is the total number of time instants in the fourth preset time period.

[0039] Step 1012, obtaining the daily average target time interval for calculating the power prediction deviation of the photovoltaic power station according to the average actual power of each time instant in the third preset time period and the fourth preset time period of the daily average sunshine duration interval.

[0040] In one embodiment of the present application, the method for obtaining the daily average target time interval for calculating the power prediction deviation of the photovoltaic power station according to the average actual power of each time in the third preset time interval and the fourth preset time interval of the daily average sunshine time interval can include the following steps:

[0041] Step a, P i is compared with the power threshold, wherein i = n;

[0042] Step b, if P i is greater than or equal to the power threshold, P n-1 Step a is repeated, and when P1 is greater than or equal to the power threshold, all times in the third preset time interval are not processed.

[0043] Step c, if P i is less than the power threshold, the i time and all times before the i time are deleted, the daily average target time interval starts from the i+1 time, and when P n is less than the power threshold, the daily average target time interval starts from the time after the third preset time interval.

[0044] Step d, M j is compared with the power threshold, wherein j = 1;

[0045] Step e, if M j is greater than or equal to the power threshold, M j+1 Step d is repeated, and when M m is greater than or equal to the power threshold, all times in the fourth preset time interval are not processed.

[0046] Step f, if M j is less than the power threshold, the j time and all times after the j time are deleted, the daily average target time interval ends at the j-1 time, and when M1 is less than the power threshold, the daily average target time interval ends at the time before the fourth preset time interval.

[0047] In one embodiment of the present application, the above power threshold = βC, wherein β can be set according to the current average sunshine of the station, and the value range of β is [0.05, 0.1], and C is the rated capacity of the photovoltaic station.

[0048] For example, in one embodiment of the present application, it is assumed that the third preset time period is [6:00, 8:00), the fourth preset time period is (16:00, 18:00], and one time period is 15 minutes. P4 and M2 are determined to be less than the power threshold value by the above method. Based on this, it can be determined that the daily average target time interval starts from the fifth time period of the third preset time period and ends at the first time period of the fourth time period, i.e., the daily average target time interval starts from 7:00 and ends at 16:15, so the daily average target time interval is [7:00, 16:15].

[0049] In addition, in one embodiment of the present application, the above case two can determine the daily average target time interval corresponding to each season by the above method.

[0050] Step 102, obtaining the actual power data and the power prediction data of each time period of the daily average target time interval in the second preset time period of the photovoltaic power station.

[0051] In one embodiment of the present application, the above second preset time period can be the past 1 day or the past 20 days, and the second preset time period can be set as needed.

[0052] Step 103, calculating the deviation index of the power prediction of the photovoltaic power station according to the power prediction data and the actual power data of each time period of the daily average target time interval in the second preset time period.

[0053] In one embodiment of the present application, the method for calculating the deviation index of the power prediction of the photovoltaic power station according to the power prediction data and the actual power data of each time period of the daily average target time interval in the second preset time period can include the following steps:

[0054] Step 1031, calculating the deviation index of the power prediction of the photovoltaic power station in each day in the second preset time period according to the power prediction data and the actual power data of each time period of the daily average target time interval in the second preset time period.

[0055] The above deviation index can include a correlation deviation value, a MAPE (Mean Absolute Percentage Error), and a deviation qualification rate index.

[0056] In addition, in one embodiment of the present application, when the indexes included in the deviation index are different, the corresponding calculation methods are also different.

[0057] Specifically, in one embodiment of the present application, when the deviation index comprises a correlation deviation value, the method for calculating the deviation index of power prediction of the photovoltaic power station each day in the second preset time period according to the power prediction data and the actual power data at each time of the daily average target time interval in the second preset time period can comprise: calculating the correlation deviation value of power prediction of the photovoltaic power station each day according to the power prediction data and the actual power data of the daily average target time interval in the second preset time period by a first formula, wherein the first formula is:

[0058]

[0059] wherein ρ A,F represents the correlation deviation value of power prediction of the photovoltaic power station, cov(A, F) represents the covariance of the power prediction data and the actual power data of the daily average target time interval, σ A represents the variance of the power prediction data of the daily average target time interval, σ F represents the variance of the actual power data of the daily average target time interval.

[0060] In addition, in one embodiment of the present application, in the above first formula,

[0061]

[0062]

[0063]

[0064]

[0065] wherein n is the total number of times of the daily average target time interval, A i is the actual power data at the i th time of the daily average target time interval, F i represents the power prediction data at the i th time of the daily average target time interval, E(A) represents the mean of the actual power data of the daily average target time interval, σ A 2 represents the covariance of the actual power data in the preset time period.

[0066] In addition, in one embodiment of the present application, the above correlation deviation value can reflect the correlation of two variables. Specifically, the stronger the correlation of the two variables, the more one variable reflects the information about the other variable.

[0067] Further, in an embodiment of the present application, the correlation deviation value of the actual power data and the predicted power data of the daily average target time interval of the photovoltaic power station can be obtained through the above steps, and the greater the obtained correlation deviation value is, the stronger the correlation between the actual power data and the predicted power data is, thereby indicating that the predicted power data is closer to the actual power data, and further indicating that the accuracy of the predicted power data is higher.

[0068] In another embodiment of the present application, when the deviation index includes MAPE, the method for calculating the deviation index of the daily power prediction of the photovoltaic power station in the second preset time period according to the power prediction data and the actual power data of each time of the daily average target time interval in the second preset time period can include: according to the power prediction data and the actual power data of the daily average target time interval of each day in the second preset time period, the MAPE of the daily power prediction of the photovoltaic power station is calculated through a second formula, wherein the second formula is:

[0069]

[0070] wherein n represents the total number of times of the daily average target time interval, A i represents the actual power data of the i th time of the daily average target time interval, F i represents the power prediction data of the i th time of the daily average target time interval, and ε represents a parameter value.

[0071] It should be noted that in an embodiment of the present application, MAPE can be used to evaluate the prediction performance, but the existing MAPE corresponds to the formula:

[0072]

[0073] wherein A i represents the actual value of the i th time, F i represents the prediction value of the i th time, and n represents the total number of times. The above existing MAPE corresponding formula does not consider the case that A i is 0. However, in actual industrial application scenarios, there may be a large number of 0 in the actual value set, such as in the example of actual power and predicted power, there is a high possibility that the actual power has a large number of 0 values, thereby the above existing MAPE formula cannot be used for calculation. The second formula above processes the denominator when calculating MAPE, thereby the second formula above can be applied to the scenario where the actual power is 0.

[0074] In addition, in an embodiment of the present application, ε in the second formula above can be artificially set according to actual conditions. For example, ε = 10 -9 Since the value of ε is very small, based on this, when A iWhen the non-zero value, the influence of epsilon on the calculation of MAPE can be ignored, at this time the second formula can be considered equivalent to the calculation formula of MAPE, and when A i When the zero value, epsilon makes the denominator not zero, thereby avoiding the problem of expression failure caused by the denominator of the formula being zero, so that the above-mentioned second formula can be applied to a variety of practical application scenarios, and the application range is wider.

[0075] In another embodiment of the present application, when the deviation index includes the deviation pass rate index, the method for calculating the deviation index of power prediction of the photovoltaic power station every day in the second preset time period according to the power prediction data and the actual power data at each time of the daily average target time interval in the second preset time period can include: according to the power prediction data and the actual power data of the daily average target time interval in the second preset time period, the deviation pass rate index of power prediction of the photovoltaic power station every day is calculated by a third formula, wherein the third formula is:

[0076]

[0077] Wherein, n represents the total number of times of the daily average target time interval, Q i represents the deviation pass rate of the power prediction data and the actual power data at the i th time of the daily average target time interval.

[0078] Wherein, in an embodiment of the present application, the calculation formula of the above-mentioned Q i is specifically:

[0079]

[0080] Wherein, A i represents the actual power data at the i th time in the daily average target time interval, F i represents the power prediction data at the i th time in the daily average target time interval.

[0081] And, in an embodiment of the present application, when the deviation rate of the power prediction data and the actual power data at the i th time in the calculation formula of the above-mentioned Q i is greater than the deviation rate threshold (such as 0.1), it is considered that the deviation rate of the power prediction data at the i th time is too large and cannot be used directly.

[0082] Further, in one embodiment of the present application, the deviation qualified rate index calculated by the third formula can reflect the proportion of the power prediction data less than or equal to the deviation rate threshold in the total power prediction data in the daily average target time interval. The smaller the deviation qualified rate calculated by the third formula is, the greater the proportion of the power prediction data less than or equal to the deviation rate threshold in the daily average target time interval is, and the more the time of the power prediction data less than the deviation rate threshold in the daily average target time interval is, thereby indicating that the overall prediction trend of the power prediction data is good and the deviation is small. Moreover, the deviation qualified rate obtained by calculation excludes human subjective factors and is objective and accurate.

[0083] Step 1032, obtaining the deviation index of the power prediction of the photovoltaic power station in the second preset time period by averaging the deviation index of the power prediction of the photovoltaic power station in each day in the second preset time period.

[0084] For example, in one embodiment of the present application, it is assumed that the number of days in the second preset time period is 10 days, and the deviation index includes the deviation qualified rate index. The deviation qualified rate indexes in the second preset time period calculated by the above step 1031 are R1, R2, R3, …, R10 respectively. 10 The deviation qualified rate index of the power prediction of the photovoltaic power station is

[0085] It should be noted that in one embodiment of the present application, the above case two needs to confirm the season to which each day in the second preset time period belongs when calculating the deviation index of the power prediction of the photovoltaic power station, and then calculates the deviation index of the power prediction of the photovoltaic power station according to the daily average target time interval corresponding to the season by the above step 1031-step 1032.

[0086] Step 104, comprehensively evaluating the deviation of the power prediction of the photovoltaic power station by the deviation index.

[0087] In one embodiment of the present application, the method of comprehensively evaluating the deviation of the power prediction of the photovoltaic power station by the deviation index can include: comprehensively evaluating the deviation of the power prediction of the photovoltaic power station by the calculation result of the fourth formula according to the mean value of the correlation deviation value, the mean value of the MAPE, and the mean value of the deviation accuracy rate index, wherein the fourth formula is:

[0088]

[0089] wherein, the mean value of the correlation deviation value is represented by the mean value of the weight coefficient of the MAPE is represented by The mean value of the deviation qualified rate index, w1 represents the weight coefficient of the correlation deviation value, w2 represents the weight coefficient of the MAPE, w3 represents the weight coefficient of the deviation qualified rate index, and w1+w2+w3=1.

[0090] In one embodiment of the present application, the fourth formula is the absolute value of the correlation deviation value of the power prediction data and the actual power data. When the deviation of the power prediction data and the actual power data is smaller, the absolute value of the corresponding correlation deviation value is closer to 1, so that the result of is closer to 0; when is smaller, it indicates that the overall deviation of the power prediction data and the actual power data is smaller; when is smaller, it indicates that the proportion of the power prediction data at each time in the daily average target time interval that is less than or equal to the deviation rate threshold is larger, which means that there are more times in the daily average target time interval that the power prediction data is less than the deviation rate threshold, and thus the overall prediction trend of the power prediction data is better and the deviation is smaller. Based on this, it can be concluded that the smaller the value of E is, the smaller the deviation of the power prediction data and the actual power data of the photovoltaic power station is, and thus the higher the accuracy of the power prediction of the photovoltaic power station is.

[0091] In one embodiment of the present application, the method for comprehensively evaluating the prediction deviation of the photovoltaic power station through the calculation result of the fourth formula can include, when the calculation result E of the fourth formula is greater than a preset threshold value, it can be determined that the deviation of the power prediction of the photovoltaic power station is larger, and the photovoltaic power station needs to be required to make rectification, so as to promote the photovoltaic power station owner to continuously improve the prediction accuracy of the photovoltaic power station and provide a reference for the power grid to determine the power generation plan.

[0092] In summary, in the comprehensive evaluation method for power prediction deviation of a photovoltaic power station provided in the application, the actual power data at each time in the daily average sunshine time interval in the first preset time period of the photovoltaic power station is obtained to obtain a daily average target time interval for calculating the power prediction deviation of the photovoltaic power station, the actual power data and the power prediction data at each time in the daily average target time interval in the second preset time period of the photovoltaic power station are obtained, the deviation index of the power prediction of the photovoltaic power station is calculated according to the power prediction data and the actual power data at each time in the daily average target time interval in the second preset time period, and the power prediction deviation of the photovoltaic power station is comprehensively evaluated through the deviation index. In the application, the actual power data at each time in the daily average sunshine time interval in the first preset time period of the photovoltaic power station is obtained, the time with weak light in the daily average sunshine time interval and large deviation is deleted to obtain the daily average target time interval for calculating the power prediction deviation of the photovoltaic power station, and the deviation index of the photovoltaic power station is calculated by using the actual power data and the power prediction data at each time in the daily average target time interval, so that the result of the comprehensive evaluation of the power prediction deviation of the photovoltaic power station is more accurate, and the result of the comprehensive evaluation can be used to compare the advantages and disadvantages of different prediction results.

[0093] Embodiment two

[0094] Figure 2 The structure diagram of the comprehensive evaluation device for power prediction deviation of a photovoltaic power station provided according to an embodiment of the application is shown in Figure 2 The device can include:

[0095] The first calculation module 201 is configured to obtain the daily average target time interval for calculating the power prediction deviation of the photovoltaic power station by using the actual power data at each time in the daily average sunshine time interval in the first preset time period of the photovoltaic power station.

[0096] The acquisition module 202 is configured to obtain the actual power data and the power prediction data at each time in the daily average target time interval in the second preset time period of the photovoltaic power station.

[0097] The second calculation module 203 is configured to calculate the deviation index of the power prediction of the photovoltaic power station according to the power prediction data and the actual power data at each time in the daily average target time interval in the second preset time period.

[0098] The evaluation module 204 is configured to comprehensively evaluate the power prediction deviation of the photovoltaic power station through the deviation index.

[0099] In summary, the photovoltaic power station power prediction deviation comprehensive evaluation device provided in the application, by obtaining the actual power data of each time interval of the daily average sunshine time interval in the first preset time period of the photovoltaic power station, the daily average target time interval for calculating the power prediction deviation of the photovoltaic power station is obtained, the actual power data and the power prediction data of each time interval of the daily average target time interval in the second preset time period of the photovoltaic power station are obtained, and the deviation index of the power prediction of the photovoltaic power station is calculated according to the power prediction data and the actual power data of each time interval of the daily average target time interval in the second preset time period, and the deviation index is used to comprehensively evaluate the power prediction deviation of the photovoltaic power station. In the application, the actual power data of each time interval of the daily average sunshine time interval in the first preset time period of the photovoltaic power station is obtained, the time interval with weak light in the daily average sunshine time interval is deleted, the daily average target time interval for calculating the power prediction deviation of the photovoltaic power station is obtained, and the deviation index of the photovoltaic power station is calculated by using the actual power data and the power prediction data of each time interval in the daily average target time, so that the result of the comprehensive evaluation of the power prediction deviation of the photovoltaic power station is more accurate, and the result of the comprehensive evaluation can be used to compare the advantages and disadvantages of different prediction results.

[0100] In order to realize the above-mentioned embodiments, the present disclosure further provides a computer storage medium.

[0101] The computer storage medium provided by the embodiments of the present disclosure stores an executable program; after the executable program is executed by a processor, the method shown in any of Figure 1 The computer storage medium provided by the embodiments of the present disclosure stores an executable program; after the executable program is executed by a processor, the method shown in any of

[0102] In order to realize the above-mentioned embodiments, the present disclosure further provides a computer device.

[0103] The computer device provided by the embodiments of the present disclosure comprises a memory, a processor and a computer program stored on the memory and executable on the processor; when the processor executes the program, the method shown in any of Figure 1 The computer device provided by the embodiments of the present disclosure comprises a memory, a processor and a computer program stored on the memory and executable on the processor; when the processor executes the program, the method shown in any of

[0104] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example" or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any suitable manner in any one or more embodiments or examples. In addition, the skilled in the art can combine and combine the different embodiments or examples described in the present specification and the features of the different embodiments or examples without contradiction.

[0105] Any processes or methods described in the flow charts or otherwise described herein can be understood as representing modules, segments, or portions of code that include one or more executable instructions for implementing the specified logical functions or steps, and the preferred embodiments of the application include additional or fewer steps, in other orders, with other functionality, in implementations of these preferred embodiments of the application. Thus, any of the steps, blocks, segments, or portions of the application can be implemented by computer program instructions. In this context, a "computer" or a "processor" can be "special- purpose" computer / processors such as an ASIC, an FPGA, a hardware implementation of a software module, or general-purpose computer / processors.

[0106] Although the embodiments of the present application have been shown and described above, it should be understood by those ordinary skilled in the art that the above embodiments are exemplary and cannot be understood as limiting the present application, and those ordinary skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the present application.

Claims

1. A comprehensive evaluation method of power prediction deviation of a photovoltaic power station, characterized in that, The method comprises: obtaining the daily average target time interval for calculating the power prediction deviation of the photovoltaic power station by the actual power data at each time in the daily average sunshine time interval in the first preset time period of the photovoltaic power station; obtaining the actual power data and the power prediction data at each time in the daily average target time interval in the second preset time period of the photovoltaic power station; calculating the deviation index of the power prediction of the photovoltaic power station according to the power prediction data and the actual power data at each time in the daily average target time interval in the second preset time period; comprehensively evaluating the power prediction deviation of the photovoltaic power station through the deviation index; wherein, the daily average target time interval for calculating the power prediction deviation of the photovoltaic power station is obtained by the actual power data at each time in the daily average sunshine time interval in the first preset time period of the photovoltaic power station, comprising: calculating the average actual power at each time in the third preset time period and the fourth preset time period in the daily average sunshine time interval according to the actual power data at each time in the daily average sunshine time interval in the first preset time period; obtaining the daily average target time interval for calculating the power prediction deviation of the photovoltaic power station according to the average actual power at each time in the third preset time period and the fourth preset time period in the daily average sunshine time interval; wherein, the deviation index of the power prediction of the photovoltaic power station is calculated according to the power prediction data and the actual power data at each time in the daily average target time interval in the second preset time period, comprising: calculating the deviation index of the power prediction of the photovoltaic power station according to the power prediction data and the actual power data at each time in the daily average target time interval in the second preset time period; averaging the deviation index of the power prediction of the photovoltaic power station in each day in the second preset time period based on the number of days in the second preset time period to obtain the deviation index of the power prediction of the photovoltaic power station; wherein, the deviation index of the power prediction of the photovoltaic power station comprises a correlation deviation value, a mean absolute percentage error (MAPE), and a deviation qualification rate index.

2. The method of claim 1, wherein, When the deviation index includes the correlation deviation value, the deviation index of the power prediction of the photovoltaic power station in each day in the second preset time period is calculated according to the power prediction data and the actual power data at each time in the daily average target time interval in the second preset time period, comprising: calculating the correlation deviation value of the power prediction of the photovoltaic power station in each day according to the power prediction data and the actual power data in the daily average target time interval in the second preset time period by a first formula, wherein the first formula is: wherein, ρ A,F cov(A, F) represents the covariance of the power prediction data and the actual power data of the daily target time interval, σ A cov(A, F) represents the covariance of the power prediction data and the actual power data of the daily target time interval, σ F cov(A, F) represents the covariance of the power prediction data and the actual power data of the daily target time interval, σ 3. The method of claim 1, wherein, When the deviation index includes the MAPE, the deviation index of the power prediction of the photovoltaic power station in each day in the second preset time period is calculated according to the power prediction data and the actual power data at each time in the daily average target time interval in the second preset time period, comprising: calculating the MAPE of the power prediction of the photovoltaic power station in each day according to the power prediction data and the actual power data in the daily average target time interval in each day in the second preset time period by a second formula, wherein the second formula is: wherein n represents the total number of time instants of the daily target time interval, A i represents the actual power data at the i-th time instant of the daily target time interval, F i represents the power prediction data at the i-th time instant of the daily target time interval, and ε represents a parameter value.

4. The method of claim 1, wherein, When the deviation index comprises a deviation pass rate index, the deviation index of the power prediction of the photovoltaic power station in each day in the second preset time period is calculated according to the power prediction data and the actual power data at each time in the daily average target time interval in the second preset time period, comprising: the deviation pass rate index of the power prediction of the photovoltaic power station in each day is calculated according to the power prediction data and the actual power data in the daily average target time interval in the second preset time period through a third formula, wherein the third formula is: Wherein, n represents the total number of time points of the daily target time interval, Q i represents the qualified rate of the deviation of the power prediction data and the actual power data at the i th time point of the daily target time interval.

5. The method of claim 1, wherein, The deviation of the power prediction of the photovoltaic power station is comprehensively evaluated according to the deviation index, comprising: the deviation of the power prediction of the photovoltaic power station is comprehensively evaluated according to the calculation result of a fourth formula according to the mean value of the correlation deviation value, the mean value of the MAPE, and the mean value of the deviation accuracy index, wherein the fourth formula is: wherein represents a mean value of the correlation deviation value, represents a mean value of the weight coefficient of the MAPE, represents a mean value of the deviation pass rate index, w1 represents a weight coefficient of the correlation deviation value, w2 represents a weight coefficient of the MAPE, w3 represents a weight coefficient of the deviation pass rate index, and w1 + w2 + w3 = 1.

6. A device for comprehensive evaluation of power prediction deviation of a photovoltaic power station, characterized in that, The device comprises: A first calculation module is configured to obtain a daily average target time interval for calculating the deviation of the power prediction of the photovoltaic power station according to the actual power data at each time in a daily average sunshine time interval in a first preset time period of the photovoltaic power station; An acquisition module is configured to acquire actual power data and power prediction data at each time in a daily average target time interval in a second preset time period of the photovoltaic power station; A second calculation module is configured to calculate a deviation index of the power prediction of the photovoltaic power station according to the power prediction data and the actual power data at each time in the daily average target time interval in the second preset time period; An evaluation module is configured to comprehensively evaluate the deviation of the power prediction of the photovoltaic power station according to the deviation index; The first calculation module is specifically configured to: Calculate the average actual power at each time in a third preset time period and a fourth preset time period in the daily average sunshine time interval according to the actual power data at each time in the daily average sunshine time interval in the first preset time period, respectively; Obtain the daily average target time interval for calculating the deviation of the power prediction of the photovoltaic power station according to the average actual power at each time in the third preset time period and the fourth preset time period in the daily average sunshine time interval; The second calculation module is specifically configured to: Calculate the deviation index of the power prediction of the photovoltaic power station in each day in the second preset time period according to the power prediction data and the actual power data at each time in the daily average target time interval in the second preset time period; Obtain the deviation index of the power prediction of the photovoltaic power station by averaging the deviation index of the power prediction of the photovoltaic power station in each day in the second preset time period based on the number of days in the second preset time period; The deviation index of the power prediction of the photovoltaic power station comprises a correlation deviation value, a mean absolute percentage error (MAPE), and a deviation pass rate index.

7. A computer device, comprising: The computer program is stored in the memory and executable on the processor, and the processor executes the program to implement the method of any one of claims 1-5.

Images