A method and system for adjusting short-term photovoltaic forecast curves
By preprocessing historical data from photovoltaic power plants and calculating adjustment coefficients, the photovoltaic power prediction model is dynamically adjusted, solving the problem of declining photovoltaic power prediction accuracy and achieving higher prediction accuracy and grid dispatch optimization.
Patent Information
- Application Number
- CN202310791101.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-30
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2043-06-30
AI Technical Summary
Existing photovoltaic power prediction algorithms suffer from decreased accuracy when weather and sunshine conditions change, making it difficult to meet the grid's demand for stable power.
By acquiring historical measured power data and short-term predicted power data of photovoltaic power plants, preprocessing the data, calculating adjustment coefficients, and using curve adjustment algorithms to adjust the short-term predicted power, the prediction model is optimized.
It improves the accuracy of photovoltaic power prediction, adapts to changes in weather and sunshine conditions, optimizes real-time grid dispatching plans, reduces reliance on historical data, and simplifies the algorithm structure.
Smart Images

Figure CN116995655B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of photovoltaic power prediction, and in particular to a method and system for adjusting short-term photovoltaic prediction curves. Background Technology
[0002] Photovoltaic power generation is a clean and renewable energy source with increasingly widespread applications. However, due to the influence of various factors such as weather and sunshine, the output power of photovoltaic power generation fluctuates significantly, making it difficult to meet the grid's demand for stable power. Therefore, achieving accurate prediction of photovoltaic power output is of great significance for improving the reliability of photovoltaic power generation and reducing the operating costs of the grid.
[0003] Typical photovoltaic power prediction algorithms usually build mathematical models based on historical data, such as time series models and machine learning models. However, these algorithms require a large amount of historical data for training, and the models are relatively complex. When real-time weather and sunshine conditions change, the prediction accuracy will decrease. Summary of the Invention
[0004] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.
[0005] In view of the problem of decreased prediction accuracy when weather conditions change abruptly or the assessment methods change, as described above and / or in existing systems, this invention is proposed.
[0006] Therefore, the problem to be solved by this invention is how to adjust the prediction model based on the results of other short-term prediction methods, so as to improve the prediction accuracy.
[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution:
[0008] In a first aspect, embodiments of the present invention provide a method for adjusting the short-term forecast curve of a photovoltaic power plant, which includes acquiring historical measured power data and short-term forecast power data of a photovoltaic power plant; preprocessing the acquired data to remove invalid data; calculating an adjustment coefficient based on the characteristics and actual needs of the photovoltaic power plant; and determining the optimal adjustment coefficient through a curve adjustment algorithm to adjust the short-term forecast power.
[0009] As a preferred embodiment of the photovoltaic short-term forecast curve adjustment method described in this invention, the preprocessing of the acquired data includes the following steps: historical measured power data of the photovoltaic power station at a certain moment. like If the installed capacity C of the photovoltaic power station is exceeded, then the measured power data at that moment will be... like and Then let the measured power data at that moment be... like The measured power data of the photovoltaic power station at that moment is considered invalid; short-term power data of the photovoltaic power station at a certain moment is considered invalid. like If the installed capacity C of the photovoltaic power station is exceeded, then the measured power data at that moment will be... If the short-term power data of a photovoltaic power station at a certain moment If so, the predicted power data of the photovoltaic power station at that moment is considered invalid.
[0010] As a preferred embodiment of the photovoltaic short-term forecast curve adjustment method of the present invention, the calculation of the adjustment coefficient includes the following steps: calculating the daily forecast data P in Π. i Maximum predicted power The adjustment value is determined by calculating the difference and ratio between the original predicted power and the maximum predicted power for a single day; the adjustment value is added to the original predicted power to calculate the adjusted power; the average accuracy rate after adjustment is calculated, and the adjustment coefficient is determined based on the average accuracy rate.
[0011] As a preferred embodiment of the photovoltaic short-term forecast curve adjustment method described in this invention, the specific formula for the adjustment value is as follows:
[0012]
[0013] in, For short-term forecast data points, For short-term forecast data points The corresponding adjustment value, k is the adjustment coefficient, P i max This represents the maximum predicted power for the day.
[0014] As a preferred embodiment of the photovoltaic short-term forecast curve adjustment method described in this invention, the specific formula for the adjusted power is as follows:
[0015]
[0016] Where, k max For the optimal adjustment factor, For short-term forecast data points, For short-term forecast data points The corresponding adjusted predicted power, P i max This represents the maximum predicted power for the day.
[0017] As a preferred embodiment of the photovoltaic short-term forecast curve adjustment method of the present invention, the method for determining the adjustment coefficient based on the average accuracy includes the following steps: traversing the adjustment coefficient k within the interval [-1, 1] according to a given step size S; calculating the prediction accuracy for each adjustment coefficient based on its corresponding predicted power data; calculating the average accuracy for each day within the time period; and calculating the adjusted average accuracy according to the specific accuracy calculation formula L, as follows:
[0018]
[0019] Among them, The adjustment factor is k i The set of time-adjusted predicted power data points The adjustment factor is k i The accuracy of the corresponding time.
[0020] Compare each of the traversed The corresponding k with the highest accuracy i The optimal adjustment coefficient k max Determine k max As an adjustment factor.
[0021] As a preferred embodiment of the photovoltaic short-term forecast curve adjustment method described in this invention, the specific formula for the forecast accuracy is as follows:
[0022]
[0023] Where Acc is the prediction accuracy, Cap is the operating capacity of the new energy power plant on the assessment day, and f i Let p be the actual power output of the new energy power station at time i. i Let i be the predicted power of the new energy power station, and n be the number of prediction points for daily assessment.
[0024] Secondly, to further address the problem of decreased prediction accuracy when weather conditions change abruptly or assessment methods change, this invention provides a photovoltaic short-term forecast curve adjustment system. This system includes a data acquisition module for acquiring historical measured power data and short-term predicted power data from photovoltaic power plants and storing them in a database; a data preprocessing module for preprocessing the historical measured power data and short-term predicted power data, removing data exceeding the installed capacity of the photovoltaic power plant and invalid data; a coefficient calculation module for calculating adjustment coefficients based on the characteristics and actual needs of the photovoltaic power plant, selecting an appropriate accuracy calculation method; a curve adjustment module for traversing the adjustment coefficients within an interval, comparing each traversed adjustment coefficient, selecting the one with the highest accuracy as the optimal adjustment coefficient, and adjusting the short-term predicted power based on the optimal adjustment coefficient and the curve adjustment algorithm; and a result output module for outputting the adjusted short-term predicted power to the user and providing a visual display.
[0025] Thirdly, embodiments of the present invention provide a computer device, including a memory and a processor, wherein the memory stores a computer program, and the computer program, when executed by the processor, implements any step of the above-described method for adjusting the short-term photovoltaic forecast curve.
[0026] Fourthly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the above-described photovoltaic short-term forecast curve adjustment method.
[0027] The beneficial effects of this invention are as follows: This invention can adaptively adjust the prediction model using recent data to adapt to changes in weather and sunshine conditions or regional assessment methods, thereby adjusting and correcting the prediction model to improve the accuracy of power prediction, optimize and adjust the real-time power grid dispatch plan, and provide effective support for power dispatching departments; it does not require a large amount of historical data for training, and can still effectively improve prediction accuracy even when the amount of data is relatively limited or incomplete; the algorithm is relatively simple to implement, does not require complex model structures or a large amount of computing resources, and is easy to implement and apply. Attached Figure Description
[0028] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:
[0029] Figure 1 This is a flowchart of the photovoltaic short-term forecast curve adjustment method in Example 1.
[0030] Figure 2 This is a comparison chart of the photovoltaic short-term forecast curve adjustment method before and after the curve adjustment in Example 2. Detailed Implementation
[0031] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.
[0032] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.
[0033] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.
[0034] This invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not adhering to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.
[0035] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.
[0036] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.
[0037] Example 1
[0038] Reference Figure 1 This is the first embodiment of the present invention, which provides a method for adjusting the short-term forecast curve of photovoltaic power, including:
[0039] S1: Obtain historical measured power data and short-term predicted power data of photovoltaic power plants.
[0040] Preferably, the acquired data includes the existing effective short-term predicted power dataset Π={P 1 ,P 2 ,...,P s} and the existing valid set of measured power data Φ={F 1 ,F 2 ,...,F t}
[0041] Specifically, Π represents the set of existing effective short-term predicted power data points. For the Pth i Let Φ be the set of valid data points for the current period, where i∈(1,s), and Φ is the set of existing valid measured power data points. For the Fth i The set of valid data points for t days, where i∈(1,t).
[0042] It should be noted that in this embodiment, a data resolution of 5 min or 15 min corresponds to n=288 or n=96 respectively, but in actual application, the specific resolution is determined by the power dispatching agency.
[0043] Specifically, when the data resolution is 5 minutes, there are 12 5-minute time intervals per hour. Since the prediction range is 24 hours, there are a total of 288 5-minute time intervals, i.e., n = 288. When the data resolution is 15 minutes, there are 4 5-minute time intervals per hour. Since the prediction time range is 24 hours, there are a total of 96 15-minute time intervals, i.e., n = 96.
[0044] S2: Preprocess the acquired data and remove invalid data.
[0045] Furthermore, the preprocessing of historical measured power data and short-term predicted power data of photovoltaic power plants includes the following steps:
[0046] S2.1: Perform data preprocessing on the historical measured power data set Φ.
[0047] Specifically, the historical measured power data of a photovoltaic power station at a certain moment. Preprocessing: The data preprocessing described here aims to correct and remove some common anomalous data in engineering applications, including the following:
[0048] S2.1.1: If the measured power data of a photovoltaic power station at a certain moment... If the installed capacity C of the photovoltaic power station is exceeded, then the measured power data at that moment will be... Where C represents the installed capacity of the photovoltaic power station.
[0049] It should be noted that in photovoltaic power plants, the installed capacity refers to the maximum power generation capacity or maximum power output designed for the power plant. If the predicted data exceeds the installed capacity, it may indicate an error in the data entry or an error in the model or method. In reality, the power generation of a photovoltaic power plant will not exceed its installed capacity. Therefore, if the predicted data exceeds the installed capacity, it is very likely an error in the data entry and needs to be corrected.
[0050] S2.1.2: If the measured power data of a photovoltaic power station at a certain moment... and δ is a preset error threshold, which sets the measured power data at that moment as...
[0051] It is important to note that photovoltaic power plants convert light energy into electrical energy, so the power generation capacity must be a non-negative number. According to the principle of photovoltaic power generation technology, the greater the light intensity, the higher the power generation capacity. Therefore, the effective prediction data should be positive. If the effective prediction data is negative, it means that the photovoltaic power plant has experienced "absorption" rather than "power generation" at a certain time period, which is inconsistent with the laws of physics. Therefore, negative prediction data is considered invalid data.
[0052] S2.1.3: If the measured power data of the photovoltaic power station at a certain moment... Then the measured power data of the photovoltaic power station at that moment is considered invalid.
[0053] It should be noted that in practical engineering applications, acquisition deviations often occur. When the measured power is 0, the acquired value may be a small negative number (the actual value is 0). Therefore, a threshold δ is set to treat small negative numbers as valid data and set them to zero, while data less than the threshold δ is considered invalid data.
[0054] S2.2: Perform data preprocessing on the short-term predicted power dataset Π.
[0055] Specifically, short-term power data of a photovoltaic power station at a certain moment. Preprocessing includes the following steps:
[0056] S2.2.1: If the short-term power data of a photovoltaic power station at a certain moment... If the installed capacity C of the photovoltaic power station is exceeded, then the measured power data at that moment will be... Where C represents the installed capacity of the photovoltaic power station.
[0057] It should be noted that in photovoltaic power plants, the installed capacity refers to the maximum power generation capacity or maximum power output designed for the power plant. If the predicted data exceeds the installed capacity, it may indicate an error in the data entry or an error in the model or method. In reality, the power generation of a photovoltaic power plant will not exceed its installed capacity. Therefore, if the predicted data exceeds the installed capacity, it is very likely an error in the data entry and needs to be corrected.
[0058] S2.2.2: If the short-term power data of a photovoltaic power station at a certain moment... If so, the predicted power data of the photovoltaic power station at that moment is considered invalid.
[0059] It is important to note that photovoltaic power plants convert light energy into electrical energy, so the power generation capacity must be a non-negative number. According to the principle of photovoltaic power generation technology, the greater the light intensity, the higher the power generation capacity. Therefore, the effective prediction data should be positive. If the effective prediction data is negative, it means that the photovoltaic power plant has experienced "absorption" rather than "power generation" at a certain time period, which is inconsistent with the laws of physics. Therefore, negative prediction data is considered invalid data.
[0060] S2.3: Delete invalid data.
[0061] It is important to note that introducing invalid (erroneous) data can affect the determination of coefficients in the following text, leading to misleading results from the model and reducing the accuracy and reliability of the prediction. Therefore, invalid data needs to be removed here.
[0062] S3: Calculate the adjustment coefficient based on the characteristics and actual needs of the photovoltaic power station.
[0063] Furthermore, calculating the adjustment coefficient based on the characteristics and actual needs of the photovoltaic power station includes the following steps:
[0064] S3.1: Calculate the daily forecast data P in Π i Maximum predicted power
[0065] Specifically, extract all valid forecast data P for each day. i ={p1,p2,...,p n}, calculate the daily forecast data P respectively i Maximum value of data The specific formula is as follows:
[0066]
[0067] Among them, P i For daily forecast data, This is the maximum value.
[0068] It should be noted that if there are no abnormalities or missing data on the day, the resolution between the predicted data and the measured data is 5 minutes or 15 minutes, corresponding to n=96 or n=288.
[0069] Specifically, when the data resolution is 5 minutes, there are 12 5-minute time intervals per hour. Since the prediction range is 24 hours, there are a total of 288 5-minute time intervals, i.e., n = 288. When the data resolution is 15 minutes, there are 4 5-minute time intervals per hour. Since the prediction time range is 24 hours, there are a total of 96 15-minute time intervals, i.e., n = 96.
[0070] S3.2: The adjustment value is determined by calculating the difference and ratio between the original predicted power and the maximum predicted power for a single day.
[0071] Specifically, the closer the predicted value is to 0 or the maximum predicted power for a single day, the smaller the adjustment value will be. At the same time, the adjustment range is controlled by the adjustment coefficient k. The larger k is, the larger the adjustment range will be.
[0072] Furthermore, the specific formula for the adjustment value is as follows:
[0073]
[0074] in, For short-term forecast data points, For short-term forecast data points The corresponding adjustment value, k is the adjustment coefficient and k∈[-1,1], P i max This represents the maximum predicted power for the day.
[0075] S3.3: Add the adjusted value to the original predicted power to calculate the adjusted power.
[0076] Furthermore, the specific formula for the adjusted power is as follows:
[0077]
[0078] Where, k max For the optimal adjustment factor, For short-term forecast data points, For short-term forecast data points The corresponding adjusted predicted power, P i max This represents the maximum predicted power for the day.
[0079] S3.4: Calculate the adjusted average accuracy and determine the adjustment factor based on the average accuracy.
[0080] Furthermore, determining the adjustment factor based on the average accuracy includes the following steps:
[0081] S3.4.1: According to the step size S, traverse and adjust the coefficient k in the interval [-1,1].
[0082] It should be noted that the step size S for short-term power prediction data is 5 min or 15 min. The 5 min step size provides higher temporal resolution, enabling more detailed capture of rapid fluctuations and instantaneous changes in power. Due to the shorter time interval, the 5 min step size also makes the detection of outliers more sensitive. Outlier data may have a significant impact on the power prediction results, so early detection and handling of outlier data is crucial. In situations with limited resources or weak processing capabilities, the 15 min step size can reduce the amount of prediction data, reduce computational load, and reduce data storage and transmission overhead. Furthermore, due to the longer time interval, the 15 min step size can smooth the power data, reducing the impact of noise and instantaneous fluctuations.
[0083] Therefore, if higher temporal resolution, more accurate prediction results, and more sensitive anomaly detection are required, a 5-minute step size can be selected. If computational efficiency, smoothing, and long-term planning are of greater importance, a 15-minute step size can be selected.
[0084] S3.4.2: For each adjustment coefficient, calculate the prediction accuracy based on its corresponding predicted power data.
[0085] Specifically, the formula for calculating prediction accuracy is as follows:
[0086]
[0087] Where Acc is the daily forecast accuracy, Cap is the operating capacity of the new energy power plant on the assessment day, and f i Let p be the actual power output of the new energy power station at time i. i Let i be the predicted power of the new energy power station, and n be the number of prediction points for daily assessment.
[0088] S3.4.3: Calculate the average accuracy for each day within the time period.
[0089] The specific formula for average accuracy is as follows:
[0090]
[0091] in, It is the average accuracy before adjustment, Acc. i Let represent the prediction accuracy for day i, and n be the total number of days.
[0092] It should be noted that the accuracy calculated here is the average accuracy before adjustment.
[0093] S3.4.4: Calculate the adjusted average accuracy for the specified time period based on the specific accuracy calculation formula L.
[0094] The specific formula is as follows:
[0095]
[0096] in, The algorithm described in steps S3.2 and S3.3 uses an adjustment coefficient of k. i The set of time-adjusted predicted power data points The adjustment factor is k i The accuracy of the corresponding time.
[0097] S3.4.5: Compare each of the traversed nodes The corresponding k with the highest accuracy i The optimal adjustment coefficient k max Determine k max As an adjustment factor.
[0098] S4: Determine the optimal adjustment coefficient through the curve adjustment algorithm to adjust the short-term predicted power.
[0099] Specifically, the adjustment value is determined by calculating the difference and ratio between the original predicted power and the maximum predicted power for a single day. The closer the predicted value is to 0 or the maximum predicted power for a single day, the smaller the adjustment value. Simultaneously, the optimal adjustment coefficient k is used. max The adjustment range is controlled to obtain the adjustment value. This adjustment value is then added to the original predicted power to calculate the adjusted power. The specific formula is as follows:
[0100]
[0101] Where, k max For the optimal adjustment factor, For short-term forecast data points, For short-term forecast data points The corresponding adjusted predicted power, P i max This represents the maximum predicted power for the day.
[0102] Furthermore, this embodiment also provides a photovoltaic short-term forecast curve adjustment system, including a data acquisition module for acquiring historical measured power data and short-term forecast power data of photovoltaic power plants and storing them in a database; a data preprocessing module for preprocessing the historical measured power data and short-term forecast power data, and removing data exceeding the installed capacity of the photovoltaic power plant and invalid data; a coefficient calculation module for calculating adjustment coefficients based on the characteristics and actual needs of the photovoltaic power plant, selecting an appropriate accuracy calculation method; a curve adjustment module for traversing the adjustment coefficients within an interval, comparing each traversed adjustment coefficient, selecting the one with the highest accuracy as the optimal adjustment coefficient, and adjusting the short-term forecast power based on the optimal adjustment coefficient and the curve adjustment algorithm; and a result output module for outputting the adjusted short-term forecast power to the user and providing a visual display.
[0103] This embodiment also provides a computer device applicable to the photovoltaic short-term forecast curve adjustment method, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the photovoltaic short-term forecast curve adjustment method proposed in the above embodiment.
[0104] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.
[0105] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, performs the following steps: acquiring historical measured power data and short-term predicted power data of a photovoltaic power station; preprocessing the acquired data to remove invalid data; calculating adjustment coefficients based on the characteristics and actual needs of the photovoltaic power station; and determining the optimal adjustment coefficients through a curve adjustment algorithm to adjust the short-term predicted power.
[0106] In summary, this invention can adaptively adjust the prediction model using recent data to adapt to changes in weather and sunshine conditions or regional assessment methods, thereby improving the accuracy of power prediction, optimizing real-time power grid dispatching plans, and providing effective support for power dispatching departments. It does not require extensive training with historical data and can still effectively improve prediction accuracy even with limited or incomplete data. The algorithm is relatively simple to implement, requiring no complex model structure or large amounts of computing resources, making it easy to implement and apply.
[0107] Example 2
[0108] Reference Figure 2 This is the second embodiment of the present invention, which provides a method for adjusting the short-term forecast curve of photovoltaics. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiment.
[0109] Specifically, the actual application case of this invention in a photovoltaic power station in Hebei Province is as follows: The installed capacity of the photovoltaic power station is 70.0MW. The measured power data and the average value of the short-term predicted power data for 17 minutes at the corresponding time from April 10 to April 26, 2023, were collected as sample data.
[0110] Furthermore, the specific formula for prediction accuracy is as follows:
[0111]
[0112] Where Acc is the daily forecast accuracy, Cap is the operating capacity of the new energy power plant on the assessment day, and f i Let p be the actual power output of the new energy power station at time i. i Let i be the predicted power of the new energy power station, and n be the number of prediction points for daily assessment.
[0113] Furthermore, after preprocessing the historical measured power data and short-term predicted power data of the photovoltaic power station in step S2 to remove invalid flag data, the coefficient is adjusted using a specific accuracy calculation method in step S3 to obtain the optimal coefficient k = 0.41. The accuracy of the short-term power generation prediction of the photovoltaic power station before and after adjustment is shown in Table 1.
[0114] Table 1. Comparison of daily accuracy before and after adjustment when adjustment factor k = 0.41
[0115] date Accuracy before adjustment Adjusted accuracy 2023-04-10 75.23% 74.49% 2023-04-11 68.80% 65.48% 2023-04-12 76.73% 77.20% 2023-04-13 88.62% 88.43% 2023-04-14 80.68% 81.64% 2023-04-15 73.21% 73.14% 2023-04-16 86.06% 89.88% 2023-04-17 70.73% 73.32% 2023-04-18 86.77% 91.64% 2023-04-19 26.54% 25.55% 2023-04-20 84.85% 85.55% 2023-04-21 84.98% 87.27% 2023-04-22 77.80% 80.93% 2023-04-23 66.04% 71.06% 2023-04-24 81.14% 81.68% 2023-04-25 67.19% 68.36% 2023-04-26 63.86% 63.35%
[0116] Furthermore, Table 1 shows the daily accuracy before and after adjustment when the adjustment factor k = 0.41 in the 17-day example data, and Table 2 shows the average accuracy before and after adjustment when the adjustment factor k = 0.41.
[0117] Table 2 Comparison of average accuracy before and after adjustment when adjustment factor k = 0.41
[0118] Accuracy before adjustment Adjusted accuracy average value 74.07% 75.32%
[0119] Preferably, as can be seen from Tables 1 and 2, in the 17-day example data, the adjusted accuracy was higher than the unadjusted accuracy on 11 days, and the average of the adjusted accuracy was 1.25% higher than the average of the unadjusted accuracy. The above data shows that the short-term prediction curve adjustment algorithm can effectively improve the accuracy of the short-term prediction model.
[0120] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
Claims
1. A method for adjusting the short-term forecast curve of photovoltaic power, characterized in that: include, Obtain historical measured power data and short-term predicted power data of photovoltaic power plants; The acquired data is preprocessed to remove invalid data; Calculate the adjustment coefficient based on the characteristics and actual needs of the photovoltaic power station; The calculation of the adjustment factor includes the following steps: Calculate the short-term forecast power dataset Daily forecast data Maximum predicted power ; The adjustment value is determined by calculating the difference and ratio between the original predicted power and the maximum predicted power for a single day; Add the adjusted value to the original predicted power to calculate the adjusted power; Calculate the adjusted average accuracy and determine the adjustment factor based on the average accuracy. The specific formula for the adjustment value is as follows: in, For the short-term forecast data point at time j on day i, For short-term forecast data points The corresponding adjustment value, To adjust the coefficient, This represents the maximum predicted power for the day. The optimal adjustment coefficient is determined by the curve adjustment algorithm, and the short-term predicted power is adjusted accordingly. Determining the adjustment factor based on the average accuracy includes the following steps: According to the given step size In the interval Inner traversal adjustment coefficient ; For each adjustment factor, the prediction accuracy is calculated based on its corresponding predicted power data. Calculate the average accuracy for each day within the time period; According to the specific accuracy calculation formula Calculate the adjusted average accuracy using the following formula: Among them, The adjustment factor is The adjusted predicted power data point set at that time The adjustment factor is The accuracy of the corresponding time, This is a collection of historical measured power data; Compare each of the traversed The one with the highest accuracy For optimal adjustment coefficient ,Sure As an adjustment factor.
2. The photovoltaic short-term forecast curve adjustment method as described in claim 1, characterized in that: The preprocessing of the acquired data includes the following steps: Historical measured power data of a photovoltaic power station at a certain moment : like Exceeding the installed capacity of photovoltaic power plants Then let the measured power data at that moment be... ; like and Then let the measured power data at that moment be... ; like If the measured power data of the photovoltaic power station at that moment is invalid, then the measured power data of the photovoltaic power station at that moment is considered invalid. Short-term power data of a photovoltaic power station at a certain moment : like Exceeding the installed capacity of photovoltaic power plants Then let the measured power data at that moment be... ; If the short-term power data of a photovoltaic power station at a certain moment If the predicted power data of the photovoltaic power station at that moment is invalid, then the predicted power data of the photovoltaic power station at that moment is considered invalid.
3. The photovoltaic short-term forecast curve adjustment method as described in claim 1, characterized in that: The specific formula for the adjusted power is as follows: in, For the optimal adjustment factor, For short-term forecast data points, For short-term forecast data points The corresponding adjusted predicted power, This represents the maximum predicted power for the day.
4. The photovoltaic short-term forecast curve adjustment method as described in claim 1, characterized in that: The specific formula for the prediction accuracy is as follows: in, To improve prediction accuracy, The operating capacity of the new energy power plant on the assessment day. for j Real-time power output of new energy power plants for j The predicted power output of new energy power plants at any given time, where n is the number of prediction points assessed daily.
5. A photovoltaic short-term forecast curve adjustment system, based on the photovoltaic short-term forecast curve adjustment method according to any one of claims 1 to 4, characterized in that: It also includes, The data acquisition module is used to acquire historical measured power data and short-term predicted power data of photovoltaic power plants and store them in the database; The data preprocessing module is used to preprocess historical measured power data and short-term predicted power data, and to remove data that exceeds the installed capacity of photovoltaic power plants and invalid data. The coefficient calculation module is used to calculate the adjustment coefficient by selecting an appropriate accuracy calculation method based on the characteristics and actual needs of the photovoltaic power station. The curve adjustment module is used to traverse the adjustment coefficients within the interval, compare each traversed adjustment coefficient, select the one with the highest accuracy as the optimal adjustment coefficient, and adjust the short-term predicted power based on the optimal adjustment coefficient and the curve adjustment algorithm. The results output module outputs the adjusted short-term predicted power to the user and can be visualized.
6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the photovoltaic short-term forecast curve adjustment method according to any one of claims 1 to 4.
7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the photovoltaic short-term forecast curve adjustment method according to any one of claims 1 to 4.
Citation Information
Patent Citations
Method for optimizing and correcting wind power prediction models
CN103473438A
Photovoltaic power station generated power super-short-term prediction method
CN104268659A
