Photovoltaic power prediction method and system based on light intensity
By processing the foundation cloud map, the relationship between cloud cluster coverage and solar radiation is established, and the light intensity is corrected in combination with historical data, the problem of large calculation amount and low accuracy of photovoltaic power prediction is solved, and efficient and accurate photovoltaic power prediction is achieved.
Patent Information
- Application Number
- CN202411273491.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-12
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-09-12
AI Technical Summary
The existing photovoltaic power prediction technology has the problem of large calculation volume, high deployment difficulty and low prediction accuracy in areas with relatively stable climate and environmental changes.
By processing the foundation cloud map, the relationship between cloud cluster coverage and solar radiation is established, the light intensity is predicted, and the output power of the photovoltaic system is corrected based on historical data.
It reduces the calculation amount and deployment difficulty of photovoltaic power prediction, improves the accuracy and efficiency of prediction, and is suitable for photovoltaic systems in different climates and environments.
Smart Images

Figure CN118801376B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of photovoltaic power prediction, and in particular to a photovoltaic power prediction method and system based on light intensity. Background Art
[0002] At present, my country's photovoltaic power generation system is mainly a DC system, that is, the electricity generated by the solar cell is charged to the battery, and the battery directly supplies power to the load. Therefore, the power generation of the photovoltaic module is closely related to the amount of solar radiation. In the actual operation of the photovoltaic power station, the photovoltaic power generation power changes with the change of the intensity of solar radiation. Since solar radiation is easily affected by the environment, it will have strong volatility and randomness. Therefore, photovoltaic power generation also has the characteristics of volatility, intermittency and randomness. These characteristics lead to power fluctuations in the photovoltaic field, which in turn will affect the overall operation of the regional power grid, and then affect the voltage stability in the entire regional grid. When the photovoltaic power station is connected to the power grid, it will bring certain hidden dangers to the safe and stable operation of the entire power system. Therefore, the accurate prediction of solar radiation has important guiding significance for the real-time monitoring of photovoltaic power, the grid-connected dispatching of photovoltaic systems, and the intelligent planning of photovoltaic power stations.
[0003] In the prior art, the solution for photovoltaic prediction is mainly based on complex neural networks to make predictions by obtaining changes in the external environment. However, the problems existing in the prior art are that, on the one hand, the training process of the model is relatively complicated because the training of the complex model requires a large data set and data features. On the other hand, the accuracy of the model is related to the degree of change of the external climate environment. For areas where the external climate environment changes relatively smoothly, the prediction accuracy of the model will be reduced. Summary of the invention
[0004] In order to solve the above technical problems, the present invention provides a photovoltaic power prediction method and system based on light intensity. By processing the ground-based cloud map, the relationship between cloud coverage and solar irradiation is established, thereby achieving the technical effect of reducing the calculation amount and deployment difficulty of photovoltaic power prediction and improving the accuracy of photovoltaic power prediction.
[0005] In a first aspect, the present invention provides a photovoltaic power prediction method based on light intensity, the method comprising:
[0006] According to the ground-based cloud map of the target area at the current moment, ground-based cloud map data is obtained, and according to the ground-based cloud map data, cloud coverage at a preset moment is predicted to obtain a cloud coverage prediction map;
[0007] According to the cloud coverage prediction map, the cloud proportion coefficient of the target area at the preset time is calculated to obtain a predicted value of the cloud proportion coefficient, and according to the predicted value of the cloud proportion coefficient and the pre-acquired solar theoretical irradiance value, a predicted value of the light intensity at the preset time is obtained;
[0008] The output power of the photovoltaic system in the target area is calculated according to the light intensity prediction value to obtain a photovoltaic power prediction value.
[0009] Furthermore, the step of obtaining ground-based cloud map data according to the ground-based cloud map of the target area at the current moment, and predicting the cloud coverage at a preset moment according to the ground-based cloud map data to obtain the cloud coverage prediction map includes:
[0010] Extracting data from the ground-based cloud map of the target area at the current moment to obtain ground-based cloud map data, wherein the ground-based cloud map data includes cloud amount data, cloud thickness data, and cloud movement data;
[0011] A cloud coverage map is obtained based on the cloud amount data and the cloud thickness data, and the cloud coverage at a preset time is predicted based on the cloud movement data and the cloud coverage map to obtain a cloud coverage prediction map.
[0012] Furthermore, the step of calculating the cloud proportion coefficient of the target area at a preset time according to the cloud coverage prediction map to obtain a predicted value of the cloud proportion coefficient includes:
[0013] According to the cloud coverage prediction map, the number of cloud pixels and the number of sky pixels in the target area at a preset time are obtained;
[0014] The sum of the number of cloud pixels and the number of sky pixels is taken as the total number of pixels, and the quotient of the number of cloud pixels divided by the total number of pixels is taken as the predicted value of the cloud proportion coefficient.
[0015] Furthermore, the step of obtaining the number of cloud pixels and the number of sky pixels in the target area at a preset time according to the cloud coverage prediction map includes:
[0016] Dividing the cloud coverage prediction map according to unit area to obtain a plurality of cloud coverage prediction sub-maps, and obtaining the corresponding sub-map cloud pixel number and sub-map sky pixel number according to each cloud coverage prediction sub-map;
[0017] The sum of the number of cloud pixels in each sub-image is taken as the number of cloud pixels, and the sum of the number of sky pixels in each sub-image is taken as the number of sky pixels.
[0018] Furthermore, after the step of obtaining the light intensity prediction value at the preset time, the method further includes:
[0019] The cloud coverage prediction map is divided according to unit area to obtain a plurality of cloud coverage prediction sub-maps, and according to each cloud coverage prediction sub-map, a corresponding sub-map cloud average thickness value and a sub-map cloud proportion coefficient prediction value are obtained;
[0020] According to the average thickness value of the sub-image cloud clusters and the predicted value of the sub-image cloud cluster ratio coefficient, the influence coefficient of the cloud cluster thickness corresponding to each cloud cluster coverage prediction sub-image on the light intensity is obtained, and according to the influence coefficient and the standard influence coefficient, the average influence coefficient is obtained;
[0021] The light intensity prediction value is corrected according to the average value of the influence coefficient to obtain a light intensity prediction correction value.
[0022] Furthermore, before the step of obtaining the average value of the influence coefficient according to the influence coefficient and the standard influence coefficient, the method further includes:
[0023] Obtain a historical light intensity prediction value and a historical light intensity actual value of the target area, compare the historical light intensity prediction value with the historical light intensity actual value, and correct the standard influence coefficient according to the comparison result to obtain a corrected standard influence coefficient.
[0024] Furthermore, the light intensity prediction value is expressed by the following formula:
[0025]
[0026] Where D t represents the predicted value of light intensity at the preset time t, N t represents the predicted value of the cloud ratio coefficient at the preset time t, D t max Indicates the theoretical solar irradiance value.
[0027] Furthermore, the influence coefficient is expressed by the following formula:
[0028]
[0029] In the formula, represents the influence coefficient of the ith cloud coverage prediction sub-graph, k represents the absorption coefficient of the cloud, and d i represents the average thickness of the sub-image cloud of the ith cloud coverage prediction sub-image, N t,i Represents the predicted value of the sub-image cloud ratio coefficient of the i-th cloud coverage prediction sub-image;
[0030] The average value of the influence coefficient is expressed by the following formula:
[0031]
[0032] Where n represents the number of cloud coverage prediction sub-graphs, represents the standard influence coefficient;
[0033] The light intensity prediction correction value is expressed by the following formula:
[0034]
[0035] Where D t Represents the predicted value of light intensity at the preset time t.
[0036] Furthermore, the step of calculating the output power of the photovoltaic system in the target area according to the light intensity prediction value to obtain the photovoltaic power prediction value includes:
[0037] Obtaining a power coefficient according to the light intensity prediction value and the standard light intensity value;
[0038] A photovoltaic power prediction value is obtained according to the power coefficient and the maximum photovoltaic power value of the photovoltaic system in the target area.
[0039] In a second aspect, the present invention provides a photovoltaic power prediction system based on light intensity, the system comprising:
[0040] A cloud coverage prediction module is used to obtain ground-based cloud map data based on the ground-based cloud map of the target area at the current moment, and to predict the cloud coverage at a preset moment based on the ground-based cloud map data to obtain a cloud coverage prediction map;
[0041] A light intensity prediction module is used to calculate the cloud ratio coefficient of the target area at a preset time according to the cloud coverage prediction map to obtain a predicted value of the cloud ratio coefficient, and to obtain a predicted value of the light intensity at the preset time according to the predicted value of the cloud ratio coefficient and a pre-acquired solar theoretical irradiance value;
[0042] The photovoltaic power prediction module is used to calculate the output power of the photovoltaic system in the target area according to the light intensity prediction value to obtain the photovoltaic power prediction value.
[0043] The present invention provides a photovoltaic power prediction method and system based on light intensity. Through the method, the relationship between cloud clusters and light intensity is analyzed using cloud map data provided by ground-based cloud maps, the influence of cloud cluster thickness on light intensity is considered, and the difference between historical prediction values and historical actual values is used for reasonable correction, so as to achieve accurate and efficient photovoltaic power prediction. The present invention not only overcomes the problems of difficult system deployment, excessive calculation amount and unstable prediction effect in the prior art, but also can effectively improve the efficiency and accuracy of photovoltaic power prediction of photovoltaic systems. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] Figure 1 is a schematic flow chart of a photovoltaic power prediction method based on light intensity in an embodiment of the present invention;
[0045] Figure 2 It is a structural schematic diagram of a photovoltaic power prediction system based on light intensity in an embodiment of the present invention. DETAILED DESCRIPTION
[0046] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0047] See also Figure 1 The first embodiment of the present invention proposes a photovoltaic power prediction method based on light intensity, which includes steps S10 to S30:
[0048] Step S10, obtaining ground-based cloud map data according to the ground-based cloud map of the target area at the current moment, and predicting the cloud coverage at a preset moment according to the ground-based cloud map data to obtain a cloud coverage prediction map;
[0049] Step S20, calculating the cloud ratio coefficient of the target area at a preset time according to the cloud coverage prediction map to obtain a predicted value of the cloud ratio coefficient, and obtaining a predicted value of the light intensity at the preset time according to the predicted value of the cloud ratio coefficient and a pre-acquired solar theoretical irradiance value;
[0050] Step S30, calculating the output power of the photovoltaic system in the target area according to the light intensity prediction value to obtain a photovoltaic power prediction value.
[0051] The present invention uses ground-based cloud maps to analyze the relationship between cloud coverage and solar irradiation to predict the light intensity, thereby calculating the photovoltaic output power of the photovoltaic system. Wherein, the ground-based cloud map refers to the cloud map obtained by ground observation equipment (such as a ground-based cloud map imager). The ground-based cloud map can intuitively describe low clouds and regional clouds, and specifically describe the external morphological characteristics of the clouds. Therefore, it is used to monitor and analyze the distribution and changes of clouds in the sky. In this embodiment, the cloud coverage prediction of the target area is achieved by analyzing the ground-based cloud map of the target area. The specific steps include:
[0052] Extracting data from the ground-based cloud map of the target area at the current moment to obtain ground-based cloud map data, wherein the ground-based cloud map data includes cloud amount data, cloud thickness data, and cloud movement data;
[0053] A cloud coverage map is obtained based on the cloud amount data and the cloud thickness data, and the cloud coverage at a preset time is predicted based on the cloud movement data and the cloud coverage map to obtain a cloud coverage prediction map.
[0054] In this embodiment, the acquired ground-based cloud image is firstly processed, and the sky image and the cloud image are divided from the ground-based cloud image by an image segmentation algorithm. In order to improve the accuracy of image segmentation, the image may be preprocessed before the image is divided.
[0055] In a preferred embodiment, the image preprocessing step includes: converting the ground-based cloud image from an RGB image to a grayscale image to reduce the complexity and data volume of image processing; performing histogram equalization on the grayscale image, using an edge detection algorithm to detect the edge of the grayscale image, and judging whether to apply different Gaussian blur parameters for denoising based on the average value of the edge detection result, thereby optimizing the image quality and ensuring the accuracy of pixel classification of cloud and non-cloud areas in the image.
[0056] After the ground-based cloud map is segmented, the cloud map is analyzed through an image processing model to estimate the cloud amount and thickness of the cloud cluster, and the motion data of the cloud cluster is analyzed by performing image registration and feature tracking on the continuous observation images of the ground-based cloud map, where the motion data of the cloud cluster includes the movement trajectory and movement speed of the cloud.
[0057] Furthermore, in order to improve the accuracy of cloud data, detection instruments can be combined to obtain ground-based cloud map data. For example, cloud detection instruments on satellites can be used to measure the vertical distribution and thickness of clouds, or ground cloud detectors and high-altitude meteorological detection instruments can be used to obtain vertical structure data of clouds. The specific method of extracting ground-based cloud map data can be flexibly selected according to actual conditions, and no excessive restrictions are made here.
[0058] After obtaining the ground-based cloud map data, a cloud map coverage map can be drawn based on the ground-based cloud map data. Specifically, a cloud coverage map at the current time can be drawn based on the cloud amount data and cloud thickness data. The cloud coverage map can reflect the size of the cloud through a proportional map and reflect the thickness of the cloud through the depth of color. Then, a cloud coverage prediction map is drawn based on the cloud movement data and the cloud coverage map at the current time. That is, based on the cloud coverage map at the current time, a cloud coverage map for the future time is drawn according to the movement trajectory and movement speed of the cloud, as a cloud coverage prediction map at a preset time.
[0059] After obtaining the cloud coverage prediction map, the cloud coverage prediction map is processed, and the cloud ratio coefficient of the target area at a preset time is calculated based on the pixel points. The specific steps include:
[0060] According to the cloud coverage prediction map, the number of cloud pixels and the number of sky pixels in the target area at a preset time are obtained;
[0061] The sum of the number of cloud pixels and the number of sky pixels is taken as the total number of pixels, and the quotient of the number of cloud pixels divided by the total number of pixels is taken as the predicted value of the cloud proportion coefficient.
[0062] In this embodiment, the sky area and the cloud area in the cloud coverage prediction map are first divided, and then the number of sky pixels in the sky area and the number of cloud pixels in the cloud area are counted respectively, and the sum of the two is taken as the total number of pixels. Finally, the quotient of the number of cloud pixels divided by the total number of pixels is taken as the predicted value of the cloud ratio coefficient, that is, the predicted value of the cloud ratio coefficient is calculated using the following formula:
[0063]
[0064] Where N t represents the predicted value of the cloud ratio coefficient at the preset time t, S C and S S Represent the number of cloud pixels and sky pixels respectively.
[0065] In order to make the statistics of the number of pixels of the sky and clouds more accurate, in a preferred embodiment, the present invention divides the cloud coverage prediction map and analyzes the sub-maps to improve the accuracy of the sky and cloud division. The specific steps include:
[0066] Dividing the cloud coverage prediction map according to unit area to obtain a plurality of cloud coverage prediction sub-maps, and obtaining the corresponding sub-map cloud pixel number and sub-map sky pixel number according to each cloud coverage prediction sub-map;
[0067] The sum of the number of cloud pixels in each sub-image is taken as the number of cloud pixels, and the sum of the number of sky pixels in each sub-image is taken as the number of sky pixels.
[0068] In this embodiment, the cloud coverage prediction map is divided into a plurality of cloud coverage prediction sub-maps according to a preset unit area, and then each sub-map is divided into sky and cloud, and the number of sky pixels and the number of cloud pixels in each sub-map are counted according to the division result, and finally the number of sky pixels of each sub-map is added to obtain the number of sky pixels of the cloud coverage prediction map, and the number of cloud pixels of each sub-map is added to obtain the number of cloud pixels of the cloud coverage prediction map, that is:
[0069]
[0070] Where n represents the number of cloud coverage prediction sub-graphs, S C,i and S S,iThey respectively represent the number of sub-image cloud pixels and the number of sub-image sky pixels corresponding to the i-th cloud cover prediction sub-image.
[0071] This embodiment divides the cloud coverage prediction map into multiple sub-images of unit area, and improves the accuracy of the sky area and cloud area division by distinguishing the sky and clouds in images with smaller areas, thereby improving the statistical accuracy of the pixels of the sky and clouds, and further improving the prediction accuracy of the cloud proportion coefficient prediction value.
[0072] After obtaining the predicted value of the cloud ratio coefficient, the predicted value of the light intensity at the preset time is calculated based on the theoretical solar irradiance value. The specific calculation formula is:
[0073]
[0074] Where D t represents the predicted value of light intensity at the preset time t, N t represents the predicted value of the cloud ratio coefficient at the preset time t, D t max Indicates the theoretical solar irradiance value.
[0075] After obtaining the predicted value of light intensity, the output power of the photovoltaic system can be calculated by following the steps below:
[0076] Obtaining a power coefficient according to the light intensity prediction value and the standard light intensity value;
[0077] A photovoltaic power prediction value is obtained according to the power coefficient and the maximum photovoltaic power value of the photovoltaic system in the target area.
[0078] In this embodiment, the ratio of the light intensity prediction value to the standard light intensity value is used as the power coefficient, and the product of the power coefficient and the maximum photovoltaic power of the photovoltaic system is used as the predicted value of the output power of the photovoltaic system. Here, the standard light intensity value and the maximum photovoltaic power are both known data that can be actually obtained. In this embodiment, it is only necessary to calculate the light intensity prediction value to achieve accurate prediction of the photovoltaic power of the photovoltaic system.
[0079] In a preferred embodiment, the present invention further provides a method for correcting the currently predicted light intensity prediction value to further improve the accuracy of the light intensity prediction value. The specific correction steps include:
[0080] The cloud coverage prediction map is divided according to unit area to obtain a plurality of cloud coverage prediction sub-maps, and according to each cloud coverage prediction sub-map, a corresponding sub-map cloud average thickness value and a sub-map cloud proportion coefficient prediction value are obtained;
[0081] According to the average thickness value of the sub-image cloud clusters and the predicted value of the sub-image cloud cluster ratio coefficient, the influence coefficient of the cloud cluster thickness corresponding to each cloud cluster coverage prediction sub-image on the light intensity is obtained, and according to the influence coefficient and the standard influence coefficient, the average influence coefficient is obtained;
[0082] The light intensity prediction value is corrected according to the average value of the influence coefficient to obtain a light intensity prediction correction value.
[0083] In the previous embodiment, the light intensity prediction value is mainly related to the coverage of clouds. In fact, the light intensity is not only related to the coverage of the sky by clouds, but also has a certain correlation with the thickness of clouds. In order to make a more accurate prediction of the light intensity, in this embodiment, the light intensity is corrected based on the thickness of clouds. Specifically, the cloud coverage prediction map is first divided according to the unit area to obtain a number of cloud coverage prediction sub-maps, and then according to the steps in the previous embodiment, each sub-map is used as a unit and the sub-map cloud ratio coefficient prediction value is calculated according to the number of pixels. At the same time, image analysis or image analysis combined with instrument measurement is used to obtain the average thickness of clouds in each sub-map. Then, according to the average cloud thickness value and the sub-map cloud ratio coefficient prediction value, the influence of cloud thickness on light intensity in each sub-map is analyzed. In this embodiment, the influence coefficient is used to characterize the influence of cloud thickness on light intensity. The specific formula of the influence coefficient is as follows:
[0084]
[0085] In the formula, represents the influence coefficient of the ith cloud cover prediction sub-graph, d i represents the average thickness of the sub-image cloud of the ith cloud coverage prediction sub-image, N t,i represents the predicted value of the sub-image cloud proportion coefficient of the ith cloud coverage prediction sub-image, k represents the absorption coefficient of the cloud, wherein the absorption coefficient k can be determined by simulating the propagation of light in the atmosphere and the absorption characteristics of the cloud layer using a numerical model such as MODTRAN.
[0086] Since the coverage of clouds per unit area may vary greatly, in order to improve the efficiency and accuracy of light intensity prediction, in this embodiment, the average value of the influence coefficient of cloud thickness on light intensity in each sub-graph is used as the influence coefficient of cloud thickness on light intensity in the entire target area. The specific calculation formula of the average value of the influence coefficient is as follows:
[0087]
[0088] Where n represents the number of cloud coverage prediction sub-graphs, Represents the standard influence coefficient, which is the influence coefficient of the cloud thickness on the light intensity in the current area, which is preset and can be determined based on the analysis of historical light data.
[0089] Then, the light intensity prediction value is corrected according to the following formula based on the average value of the influence coefficient, thereby obtaining the light intensity prediction correction value:
[0090]
[0091] Where D t Represents the predicted value of light intensity at the preset time t.
[0092] In this embodiment, in addition to predicting the light intensity based on the cloud coverage, the effect of the cloud thickness on the light intensity is also taken into consideration to improve the accuracy of the light intensity prediction.
[0093] It can be seen that the standard influence coefficient is used in the calculation formula of the above influence coefficient average value, so the standard influence coefficient will also affect the accuracy of light intensity prediction. Therefore, the present invention also provides another preferred embodiment to correct the standard influence coefficient to further improve the accuracy of light intensity prediction, and the specific steps include:
[0094] Obtain a historical light intensity prediction value and a historical light intensity actual value of the target area, compare the historical light intensity prediction value with the historical light intensity actual value, and correct the standard influence coefficient according to the comparison result to obtain a corrected standard influence coefficient.
[0095] In this embodiment, the above-mentioned light intensity prediction method is used to predict the light intensity at multiple preset time points in the target area, and a historical light intensity prediction value curve of the target area is drawn based on the prediction results. At the same time, the light intensity at these preset time points is actually measured by an instrument, and a historical light intensity actual value curve of the target area is drawn based on the measurement results.
[0096] Then, the historical light intensity prediction value curve and the historical light intensity actual value curve are compared to analyze the difference between the light intensity prediction value and the light intensity actual value. According to the comparison results of the two curves, the standard influence coefficient is corrected to make the setting of the standard influence coefficient more reasonable, which can more accurately reflect the influence of clouds on light in the area, thereby further improving the accuracy of the light intensity prediction value.
[0097] The present embodiment provides a photovoltaic power prediction method based on light intensity, which uses ground-based cloud maps to obtain cloud map data, and predicts the solar light intensity by establishing the relationship between cloud clusters and solar irradiation. It also considers the influence of cloud thickness on light intensity, and uses the gap between historical prediction values and historical actual values to make reasonable corrections, thereby improving the accuracy of light intensity prediction, and further improving the accuracy of photovoltaic power prediction based on light intensity. The present invention uses cloud map data provided by ground-based cloud maps to analyze the relationship between cloud clusters and light intensity, overcomes the problems of difficult system deployment, excessive calculation amount, and unstable prediction effect in the prior art, and effectively improves the efficiency and accuracy of photovoltaic power prediction of photovoltaic systems.
[0098] See also Figure 2 Based on the same inventive concept, a photovoltaic power prediction system based on light intensity is proposed in the second embodiment of the present invention, comprising:
[0099] The cloud coverage prediction module 10 is used to obtain ground-based cloud map data according to the ground-based cloud map of the target area at the current moment, and to predict the cloud coverage at a preset moment according to the ground-based cloud map data to obtain a cloud coverage prediction map;
[0100] The light intensity prediction module 20 is used to calculate the cloud ratio coefficient of the target area at a preset time according to the cloud coverage prediction map to obtain a predicted value of the cloud ratio coefficient, and to obtain a predicted value of the light intensity at the preset time according to the predicted value of the cloud ratio coefficient and a pre-acquired solar theoretical irradiance value;
[0101] The photovoltaic power prediction module 30 is used to calculate the output power of the photovoltaic system in the target area according to the light intensity prediction value to obtain a photovoltaic power prediction value.
[0102] The technical features and technical effects of the photovoltaic power prediction system based on light intensity proposed in the embodiment of the present invention are the same as those of the method proposed in the embodiment of the present invention, and are not described in detail here. Each module in the above-mentioned photovoltaic power prediction system based on light intensity can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
[0103] In summary, the embodiments of the present invention propose a photovoltaic power prediction method and system based on light intensity. The method obtains ground-based cloud map data according to the ground-based cloud map of the target area at the current moment, and predicts the cloud coverage at a preset moment according to the ground-based cloud map data to obtain a cloud coverage prediction map; calculates the cloud ratio coefficient of the target area at a preset moment according to the cloud coverage prediction map to obtain a cloud ratio coefficient prediction value, and obtains a light intensity prediction value at a preset moment according to the cloud ratio coefficient prediction value and the pre-acquired solar theoretical irradiance value; calculates the output power of the photovoltaic system in the target area according to the light intensity prediction value to obtain a photovoltaic power prediction value. The present invention uses the cloud map data provided by the ground-based cloud map to analyze the relationship between clouds and light intensity, overcomes the problems of difficult system deployment, excessive calculation amount and unstable prediction effect in the prior art, and effectively improves the efficiency and accuracy of photovoltaic power prediction of the photovoltaic system.
[0104] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0105] The above-mentioned embodiments only express several preferred implementation modes of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in the technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be based on the protection scope of the claims.
Claims
1. A photovoltaic power prediction method based on light intensity, characterized in that: include: According to the ground-based cloud map of the target area at the current moment, ground-based cloud map data is obtained, and according to the ground-based cloud map data, cloud coverage at a preset moment is predicted to obtain a cloud coverage prediction map; According to the cloud coverage prediction map, the cloud proportion coefficient of the target area at the preset time is calculated to obtain a predicted value of the cloud proportion coefficient, and according to the predicted value of the cloud proportion coefficient and the pre-acquired solar theoretical irradiance value, a predicted value of the light intensity at the preset time is obtained; Calculating the output power of the photovoltaic system in the target area according to the light intensity prediction value to obtain a photovoltaic power prediction value; The step of calculating the cloud ratio coefficient of the target area at a preset time according to the cloud coverage prediction map to obtain a predicted value of the cloud ratio coefficient comprises: According to the cloud coverage prediction map, the number of cloud pixels and the number of sky pixels in the target area at a preset time are obtained; The sum of the number of cloud pixels and the number of sky pixels is taken as the total number of pixels, and the quotient of the number of cloud pixels divided by the total number of pixels is taken as the predicted value of the cloud proportion coefficient; After the step of obtaining the predicted value of the light intensity at the preset time, the method further includes: The cloud coverage prediction map is divided according to unit area to obtain a plurality of cloud coverage prediction sub-maps, and according to each cloud coverage prediction sub-map, a corresponding sub-map cloud average thickness value and a sub-map cloud proportion coefficient prediction value are obtained; According to the average thickness value of the sub-image cloud clusters and the predicted value of the sub-image cloud cluster ratio coefficient, the influence coefficient of the cloud cluster thickness corresponding to each cloud cluster coverage prediction sub-image on the light intensity is obtained, and according to the influence coefficient and the standard influence coefficient, the average influence coefficient is obtained; Correcting the light intensity prediction value according to the average value of the influence coefficient to obtain a light intensity prediction correction value; The influence coefficient is expressed by the following formula: In the formula, represents the influence coefficient of the ith cloud coverage prediction sub-graph, k represents the absorption coefficient of the cloud, and d i represents the average thickness of the sub-image cloud of the ith cloud coverage prediction sub-image, N t,i Represents the predicted value of the sub-image cloud ratio coefficient of the i-th cloud coverage prediction sub-image; The average value of the influence coefficient is expressed by the following formula: Where n represents the number of cloud coverage prediction sub-graphs, represents the standard influence coefficient; The light intensity prediction correction value is expressed by the following formula: Where D t Represents the predicted value of light intensity at the preset time t.
2. The photovoltaic power prediction method based on light intensity according to claim 1, characterized in that: The step of obtaining ground-based cloud map data according to the ground-based cloud map of the target area at the current moment, and predicting the cloud coverage at a preset moment according to the ground-based cloud map data to obtain a cloud coverage prediction map comprises: Extracting data from the ground-based cloud map of the target area at the current moment to obtain ground-based cloud map data, wherein the ground-based cloud map data includes cloud amount data, cloud thickness data, and cloud movement data; A cloud coverage map is obtained based on the cloud amount data and the cloud thickness data, and the cloud coverage at a preset time is predicted based on the cloud movement data and the cloud coverage map to obtain a cloud coverage prediction map.
3. The photovoltaic power prediction method based on light intensity according to claim 1, characterized in that: The step of obtaining the number of cloud pixels and the number of sky pixels in the target area at a preset time according to the cloud coverage prediction map comprises: Dividing the cloud coverage prediction map according to unit area to obtain a plurality of cloud coverage prediction sub-maps, and obtaining the corresponding sub-map cloud pixel number and sub-map sky pixel number according to each cloud coverage prediction sub-map; The sum of the number of cloud pixels in each sub-image is taken as the number of cloud pixels, and the sum of the number of sky pixels in each sub-image is taken as the number of sky pixels.
4. The photovoltaic power prediction method based on light intensity according to claim 1, characterized in that: Before the step of obtaining the average value of the influence coefficient according to the influence coefficient and the standard influence coefficient, the step further includes: Obtain a historical light intensity prediction value and a historical light intensity actual value of the target area, compare the historical light intensity prediction value with the historical light intensity actual value, and correct the standard influence coefficient according to the comparison result to obtain a corrected standard influence coefficient.
5. The photovoltaic power prediction method based on light intensity according to claim 1, characterized in that: The light intensity prediction value is expressed by the following formula: Where D t represents the predicted value of light intensity at the preset time t, N t represents the predicted value of the cloud ratio coefficient at the preset time t, D t max Indicates the theoretical solar irradiance value.
6. The photovoltaic power prediction method based on light intensity according to claim 1, characterized in that: The step of calculating the output power of the photovoltaic system in the target area according to the light intensity prediction value to obtain the photovoltaic power prediction value comprises: Obtaining a power coefficient according to the light intensity prediction value and the standard light intensity value; A photovoltaic power prediction value is obtained according to the power coefficient and the maximum photovoltaic power value of the photovoltaic system in the target area.
7. A photovoltaic power prediction system based on light intensity, characterized in that: include: A cloud coverage prediction module is used to obtain ground-based cloud map data based on the ground-based cloud map of the target area at the current moment, and to predict the cloud coverage at a preset moment based on the ground-based cloud map data to obtain a cloud coverage prediction map; A light intensity prediction module is used to calculate the cloud ratio coefficient of the target area at a preset time according to the cloud coverage prediction map to obtain a predicted value of the cloud ratio coefficient, and to obtain a predicted value of the light intensity at the preset time according to the predicted value of the cloud ratio coefficient and a pre-acquired solar theoretical irradiance value; The step of calculating the cloud ratio coefficient of the target area at a preset time according to the cloud coverage prediction map to obtain a predicted value of the cloud ratio coefficient comprises: According to the cloud coverage prediction map, the number of cloud pixels and the number of sky pixels in the target area at a preset time are obtained; The sum of the number of cloud pixels and the number of sky pixels is taken as the total number of pixels, and the quotient of the number of cloud pixels divided by the total number of pixels is taken as the predicted value of the cloud proportion coefficient; After the step of obtaining the predicted value of the light intensity at the preset time, the method further includes: The cloud coverage prediction map is divided according to unit area to obtain a plurality of cloud coverage prediction sub-maps, and according to each cloud coverage prediction sub-map, a corresponding sub-map cloud average thickness value and a sub-map cloud proportion coefficient prediction value are obtained; According to the average thickness value of the sub-image cloud clusters and the predicted value of the sub-image cloud cluster ratio coefficient, the influence coefficient of the cloud cluster thickness corresponding to each cloud cluster coverage prediction sub-image on the light intensity is obtained, and according to the influence coefficient and the standard influence coefficient, the average influence coefficient is obtained; Correcting the light intensity prediction value according to the average value of the influence coefficient to obtain a light intensity prediction correction value; The influence coefficient is expressed by the following formula: In the formula, represents the influence coefficient of the ith cloud coverage prediction sub-graph, k represents the absorption coefficient of the cloud, and d i represents the average thickness of the sub-image cloud of the ith cloud coverage prediction sub-image, N t,i Represents the predicted value of the sub-image cloud ratio coefficient of the i-th cloud coverage prediction sub-image; The average value of the influence coefficient is expressed by the following formula: Where n represents the number of cloud coverage prediction sub-graphs, represents the standard influence coefficient; The light intensity prediction correction value is expressed by the following formula: Where D t represents the predicted value of light intensity at a preset time t; The photovoltaic power prediction module is used to calculate the output power of the photovoltaic system in the target area according to the light intensity prediction value to obtain the photovoltaic power prediction value.
